Skip to main content

Full text of "Report of the Secretary's Task Force on Black and Minority Health: Volume II: Crosscutting Issues in Minority Health"

See other formats


Volume II: 
Crosscutting Issues 
In Minority Health 



Report of the 
Secretary's Task 
Force on 



Black & 

Minority 

Health 



Margaret M. Heckler 
Secretary 



U.S. Department of Health and 
Human Services 



f^Hl<:)V99^S 



Volume II: 
Crosscutting Issues 
In Minority Health 



Report of the 
Secretary's Task 
Force on 



Black & 

Minority 
Health 



Margaret M. Heckler 
Secretary 



U.S. Department of Health and 
Human Services 

August 1985 



SECRETARY'S TASK FORCE ON BLACK AND MINORITY HEALTH 



MEMBERS 



Thomas E. Malone, Ph.D., Chairperson 
Katrina W. Johnson, Ph.D., Study Director 



Wendy Baldwin, Ph.D 
Betty Lou Dotson, J.D. 
Manning Feinleib, M.D. , Dr. 
William T. Friedewald, M.D. 
Robert Graham, M.D. 
M. Gene Handelsman 
Jane E. Henney, M.D. 
Donald R. Hopkins, M.D. 
Stephanie Lee-Miller 



P.H. 



Jaime Manzano 

J. Michael McGinnis, M.D. 

Mark Novitch, M.D. 

Clarice D. Reid, M.D. 

Everett R. Rhoades , M.D. 

William A. Robinson, M.D. , 

James L. Scott 

Robert L. Trachtenberg 

T. Franklin Williams, M.D. 



M.P.H. 



ALTERNATES 



Shirley P. Bagley, M.S. 
Claudia Baquet, M.D., M.P.H. 
Howard M. Bennett 
Cheryl Damberg, M.P.H. 
Mary Ann Danello, Ph.D. 
Jacob Feldman, Ph.D. 
Marilyn Gaston, M.D. 
George Hardy, M.D. 
John H. Kelso 



James A. Kissko 
Robert C. Kreuzburg, M.D. 
Barbara J. Lake 
Patricia L. Mackey, J.D. 
Delores Parron, Ph.D. 
Gerald H. Payne, M.D. 
Caroline I. Reuter 
Clay Simpson, Jr., Ph.D. 
Ronald J. Wylie 



VOLUME II: CROSSCUTTING ISSUES IN MINORITY HEALTH 

TABLE OF CONTENTS 

Introduction to the Task Force Report 1 

Members of the Subcommittee on Data Development 6 

PERSPECTIVES ON NATIONAL HEALTH DATA FOR MINORITIES: Report 

of the Subcommittee on Data Development 7 

Supporting Papers 

1. Benjamin S. Bradshaw, W. Parker Frisbie, Clayton W. Eifler: 
Excess and deficit mortality due to selected causes of death 
and their contribution to differences in life expectancy of 
Spanish-surnamed and other white males — 1970 and 1980 43 

2. Mary N. Haan, George A. Kaplan: The contribution of socioeconomic 
position to minority health 69 

3. M. Alfred Haynes , Girma Wolde-Tsadik, Paul Juarez: 
Associations of health problems with ethnic groups 

in ambulatory care visits 107 

4. Shiriki K. Kumanyika, Deborah L. Helitzer: Nutritional status 

and dietary patterns of racial minorities in the United States . . 118 

5. Reiko Homma True: Health care service delivery in Asian 

American communities • 193 

6. Elena S. H. Yu, Ching-Fu Chang, William T. Liu, Stephen H. Kan: 
Asian-White mortality differences: Are there excess deaths? . . . 209 

7. Elena S. H. Yu, William T. Liu, Paul Kurzeja: Physical and 

mental health status indicators for Asian-American communities . . 255 

MINORITY ACCESS TO HEALTH CARE IN THE MID-1980 's 287 

HEALTH EDUCATION AMONG MINORITY POPULATIONS 333 

MINORITY AND OTHER HEALTH PROFESSIONALS SERVING MINORITY COMMUNITIES . . 277 



111 



INTRODUCTION TO THE TASK FORCE REPORT 

Background 

The Task Force on Black and Minority Health was established by 
Secretary of Health and Human Services Margaret M. Heckler in response 
to the striking differences in health status between many minority 
populations in the United States and the nonminority population. 

In January 1984, when Secretary Heckler released the annual report 
of the Nation's health, Health, United States, 1983 , she noted that the 
health and longevity of all Americans have continued to improve, but the 
prospects for living full and healthy lives were not shared equally by 
many minority Americans. Mrs. Heckler called attention to the longstanding 
and persistent burden of death, disease, and disability experienced by 
those of Black, Hispanic, Native American, and Asian/Pacific Islander 
heritage in the United States. Among the most striking differentials 
are the gap of more than 5 years in life expectancy between Blacks and 
Whites and the infant mortality rate, which for Blacks has continued to 
be twice that of IJhites. T-Jhile the differences are particularly evident 
for Blacks, a group for whom information is most accurate, they are 
clear for Hispanics, Native Americans, and some groups of Asian/Pacific 
Islanders as well. 

By creating a special Secretarial Task Force to investigate this 
grave health discrepancy and by establishing an Office of Minority Health 
to implement the recommendations of the Task Force, Secretary Heckler 
has taken significant measures toward developing a coordinated strategy 
to improve the health status of all minority groups. 

Dr. Thomas E. Malone , Deputy Director of the National Institutes of 
Health, was appointed to head the Task Force and 18 senior DHHS executives 
whose programs affect minority health were selected to serve as primary 
members of the Task Force. \^±le many DHHS programs significantly benefit 
minority groups, the formation of this Task Force was unique in that it 
was the first time that attention was given to an integrated, comprehensive 
study of minority health concerns. 

Charge 

Secretary Heckler charged the Task Force with the following duties: 

• Study the current health status of Blacks, Hispanics, Native 
Americans, and Asian/Pacific Islanders. 

• Review their ability to gain access to and utilize the health 
care system. 

• Assess factors contributing to the long-term disparities in 
health status between the minority and nonminority populations. 



• Review existing DHHS research and service programs relative to 
minority health. 

• Recommend strategies to redirect Federal resources and programs to 
narrow the health differences between minorities and nonminorities. 

• Suggest strategies by which the public and private sectors can 
cooperate to bring about improvements in minority health. 

Approach 

After initial review of national data, the Task Force adopted a 
study approach based on the statistical technique of "excess deaths" 
to define the differences in minority health in relation to nonminority 
health. This method dramatically demonstrated the number of deaths among 
minorities that would not have occurred had mortality rates for minorities 
equalled those of nonminorities. The analysis of excess deaths revealed 
that six specific health areas accounted for more than 80 percent of the 
higher annual proportion of minority deaths. These areas are: 

• Cardiovascular and cerebrovascular diseases 

• Cancer 

• Chemical dependency 

• Diabetes 

• Homicide, suicide, and unintentional injuries 

• Infant mortality and low birthweight. 

Subcommittees were formed to explore why and to what extent these 
health differences occur and what DHHS can do to reduce the disparity. 
The subcommittees examined the most recent scientific data available 
in their specific areas and the physiological, cultural, and societal 
factors that might contribute to health problems in minority populations. 

The Task Force also investigated a number of issues that cut across 
specific health problem areas yet influence the overall health status of 
minority groups. Among those reviewed were demographic and social 
characteristics of Blacks, Hispanlcs, Native Americans, and Asian/Pacific 
Islanders; minority needs in health information and education; access to 
health care services by minorities; and an assessment of health professionals 
available to minority populations. Special analyses of mortality and 
morbidity data relevant to minority health also were developed for the 
use of Task Force. Reports on these issues appear in Volume II. 

Resources 



More than 40 scientific papers were commissioned to provide recent 
data and supplementary information to the Task Force and its subcommittees. 
Much material from the commissioned papers was incorporated into the 
subcommittee reports; others accompany the full text of the subcommittee 
reports. 



An inventory of DHHS program efforts in minority health was compiled 
by the Task Force. It includes descriptions of health care, prevention, 
and research programs sponsored by DHHS that affect minority populations. 
This is the first such compilation demonstrating the extensive efforts 
oriented toward minority health within DHHS. An index listing agencies 
and program titles appears in Volume 1. Volume VIII contains more 
detailed program descriptions as well as telephone numbers of the offices 
responsible for the administration of these programs. 

To supplement its knowledge of minority health issues , the Task 
Force communicated with individuals and organizations outside the Federal 
system. Experts in special problem areas such as data analysis, nutrition, 
or intervention activities presented up-to-date information to the Task 
Force or the subcommittees. An Hispanic consultant group provided inform- 
ation on health issues affecting Hispanics. A summary of Hispanic health 
concerns appears in Volume VIII along with an annotated bibliography of 
selected Hispanic health issues. Papers developed by an Asian/Pacific 
Islander consultant group accompany the data development report appearing 
in Volume II. 

A nationwide survey of organizations and individuals concerned with 
minority health issues was conducted. The survey requested opinions 
about factors influencing health status of minorities, examples of success- 
ful programs and suggestions for ways DHHS might better address minority 
health needs. A summary of responses and a complete listing of the 
organizations participating in the survey is included in Volume VIII. 

Task Force Report 

Volume I, the Executive Summary, includes recommendations for 
department-wide activities to improve minority health status. The 
recommendations emphasize activities through which DHHS might redirect 
its resources toward narrowing the disparity between minorities and 
nonminorities and suggest opportunities for cooperation with nonfederal 
structures to bring about improvements in minority health. Volume I 
also contains summaries of the information and data compiled by the Task 
Force to account for the health status disparity. 

Volumes II through VIII contain the complete text of the reports 
prepared by subcommittees and working groups. They provide extensive 
background information and data analyses that support the findings and 
intervention strategies proposed by the subcommittees. The reports are 
excellent reviews of research and should be regarded as state-of-the-art 
knowledge on problem areas in minority health. Many of the papers commissioned 
by the Task Force subcommittees accompany the subcommittee report. They 
should be extremely useful to those who wish to become familiar in greater 
depth with selected aspects of the issues that the Task Force analyzed. 



The full Task Force report consists of the following volumes 
Volume I: Executive Summary- 



Volume II: 



Volume III: 
Volume IV : 
Volume V: 
Volume VI: 
Volume VII: 

Volume VIII: 



Crosscutting Issues in Minority Health: 

Perspectives on National Health Data for Minorities 
Minority Access to Health Care 
Health Education and Information 

Minority and other Health Professionals Serving Minority 
Communities 

Cancer 

Cardiovascular and Cerebrovascular Diseases 

Homicide, Suicide, and Unintentional Injuries 

Infant Mortality and Low Birthweight 

Chemical Dependency 
Diabetes 

Hispanic Health Issues 

Inventory of DHHS Program Efforts in Minority Health 

Survey of the Non-Federal Community 



Perspectives On National 
Health Data For Minorities 

Report of the Subcommittee On Data 
Development 



SUBCOMMITTEE ON DATA DEVELOPMENT 



Robert Graham, M.D., Chairperson 

Assistant Surgeon General 

Administrator 

Health Resources and Services Administration 

Jane Delgado, Ph.D. 

Special Assistant on Minority Affairs 

Office of the Secretary 

Manning Feinleib, M.D. , Dr.P.H. 

Director 

National Center for Health Statistics 

Donald Hopkins, M.D. 

Deputy Director 

Centers for Disease Control 

Stephanie Lee-Miller 

Assistant Secretary for Public Health 

Office of the Secretary 

J. Michael McGinnis, M.D. 

Assistant Surgeon General 

Deputy Assistant Secretary for Health 

Director, Office of Disease Prevention and Health Promotion 

Office of the Secretary 

William Robinson, M.D. , M.P.H. 

Deputy Director 

Bureau of Health Professions 

Health Resources and Services Administration 

Staff Liaison: 
Clifford Patrick, Ph.D. 



PERSPECTIVES ON NATIONAL HEALTH DATA FOR MINORITIES 

SUMMARY 

The Data Development Subcommittee served as an preliminary review 
committee whose purpose was to guide the Task Force, provide national 
data to other subcommittees and commission reports on subjects that cut 
across all health problem areas. The subcommittee performed initial 
analyses for the Task Force by comparing the mortality between minority 
groups and the nonminority population in the United States. 

The technique of "excess deaths", a methodology particularly suited 
to the Task Force needs, allowed the Task Force to set priorities among 
health issues. This technique highlighted health needs that affect 
large proportions of the minority population. As a result of this 
technique, six causes of death were selected for more intensive 
examination. 

The approach was limited by lack of complete data for Hispanic and 
Asian/Pacific Islander populations and by its dependence on the size and 
characteristics of the nonminority comparison group. Supplementary 
efforts to identify more specifically the health status and needs of 
minorities included regional data, commissioned papers, and a range of 
other morbidity datasets. Although these sources were not more complete 
than death certificate information, they confirmed the selection of 
priority health issues. 

The papers and other reports in this volume summarize 
health-related issues that cut across each subcommittee area and reflect 
additional information reviewed by all components of the Task Force. 



PERSPECTIVES ON NATIONAL HEALTH DATA FOR MINORITIES 
FIGURES AND TABLES 



FIGURES 

Figure 1. Death Rates by Age at Death: White Males and White Females 
Average Annual Rates for All Causes of Death, U.S. 1979-81 

Figure 2. Death Rates for Black Males Relative to White Males 

Average Annual Rates for All Causes of Death, U.S., 1979-81 

Figure 3. Death Rates for Black Females Relative to White Females 

Average Annual Rates for All Causes of Death, U.S., 1979-81 

Figure 4. Death Rates for Native American Males Relative to 
White Males 
Average Annual Rates for All Causes of Death, U.S., 1979-81 

Figure 5. Death Rates for Native American Females Relative to 
White Females 
Average Annual Rates for All Causes of Death, U.S., 1979-81 

Figure 6. Death Rates for Asian Males Relative to White Males 

Average Annual Rates for All Causes of Death, U.S., 1979-81 

Figure 7. Death Rates for Asian Females Relative to White Females 

Average Annual Rates for All Causes of Death, U.S., 1979-81 



TABLES 
Table 1. 

Table 2. 



Average Annual Total and Excess Deaths in Blacks 
Selected Causes of Mortality, 1979-1981 

Life Expectancy for Males by Race at Selected Ages 
1969-71 and 1979-81 



Table 3. Life Expectancy for Females by Race at Selected Ages 
1969-71 and 1979-81 

Table 4. Relative Risk of Mortality for Blacks Compared with Whites 
All Causes of Death, 1979-81 

Table 5. Relative Risk of Mortality for Blacks Compared with Whites 
Selected Causes of Death by Age and Gender, 1979-81 

Table 6. Relative Risk of Morbidity for Blacks Compared with Whites 
Selected Conditions by Age and Gender, 1978-80 

Table 7. Percent of U.S. Population Groups Visiting a Physician 
During the Past 12 Months (by Age and by Sex) 

Table 8. Bed Disability Days Per Person During One Year 
U.S. Population Groups (by Age and by Sex) 

Table 9. Percent with Activity Limitation During the Past Year 
U.S. Population Groups (by Age and by Sex) 

Table 10. Work Days Lost Per Currently Employed Person During the Past 
Year 
U.S. Population Groups (by Age and by Sex) 

Table 11. Prevalence of End Stage Renal Disease (ESRD) by Race and 
Diagnosis 
Dialysis and Transplant Patients, 1982 



10 



PERSPECTIVES ON NATIONAL HEALTH DATA FOR MINORITIES 

OVERVIEW 

The Data Development Subcommittee of the Secretary's Task Force on 
Black and Minority Health examined a variety of methods to measure the 
disparity between the health of minority and nonminority Americans. The 
analytical approach selected by the Subcommittee was critical in guiding 
the Task Force in its efforts to establish priorities in minority 
health, structure further Task Force activities, and recommend ways to 
improve the health of Black, Hispanic, Asian/Pacific Islander, and 
Native Americans. 

Initial examination of national data as reported in Health US, 1983 
showed disparities between Black and White populations in life 
expectancy, infant mortality, several selected causes of death, and 
other indicators. While national data on the health status of Blacks are 
more complete, available data on Native Americans and Hispanics suggest 
that these populations suffer a greater disease and mortality burden 
than Whites. 

Charged with fuller exploration of health discrepancies for all 
minorities, the Task Force sought a methodology to guide their 
deliberations. After careful consideration of the data in Health US, 
1983 as well as a more detailed presentation of national datasets, the 
Task Force concluded that additional analyses were required to explore 
why the discrepancy has continued for such a long period. 

At the June 1984 meeting of the Task Force, as a result of a 
presentation by Dr. M. Alfred Haynes, President and Dean of the Drew 
Postgraduate Medical School in Los Angeles, California, Task Force 
members decided that the statistical method of "excess deaths" was 
particularly suited for comparing minority and nonminority population 
groups. At that time, the Task Force established a Data Development 
Subcommittee, charged with applying the excess deaths technique to 
national mortality data and identifying other issues to be addressed. 
Dr. Robert Graham, Administrator, Health Resources and Services 
Administration, chaired the Subcommittee. 

At the July 1984 Task Force meeting, the results of the calculation 
of excess deaths by age, sex, race, and cause of death were presented by 
the Data Development Subcommittee to the Task Force membership. The 
excess deaths analysis clearly demonstrated that six health problem 
areas most significantly affect the health of minority populations. 
These areas are: cancer, cardiovascular disease (including stroke), 
cirrhosis and liver disease (indicators of chemical dependency) , 
diabetes, homicide and unintentional injuries, and infant mortality 
(death under one year of age) . Subcommittees addressing these areas 
were established. Where national data were missing or incomplete, 
papers were commissioned reviewing regional data or special topics. 



11 



Data and analyses reviewed by the Data Development Subcoinmittee were 
provided to the six subcommittees for further exploration. 

To augment and confirm the priorities elucidated by the excess 
deaths analysis, additional measures of mortality and morbidity were 
reviewed. Additional mortality indices included "excess" person-years 
of life lost, life expectancy, and relative risk of death by cause. 
Morbidity measures included prevalence rates of selected diseases, 
hospital admissions, physician visits, limitation of activity and 
self-assessed health status. These were presented and discussed at the 
December 1984 meeting of the Task Force. A number of scientific papers 
were commissioned by the Data Development Subcommittee on special topics 
not included in other subcommittee work; these include socioeconomic 
influences on health, nutrition, health of Asian subgroups and regional 
analyses of Hispanic populations. 

The Task Force recognized that other issues cut across health 
problem areas and influence the health status of minority Americans. 
Reports representing state-of-the-art expertise in these fields were 
developed. These issues include health education, access to health care, 
and availability of health professionals to minorities. 

The report of the Data Development Subcommittee describes the 
methodology and analytical processes that led to the findings and 
recommendations reported in Volume I. The principal health indices, 
limitations in the data and methodology, and selected findings drawn 
from the methodology used by the Data Development Subcommittee are 
included. 

SUBCOMMITTEE DATA SOURCES 

The measure of excess deaths is not a routinely used mortality 
indicator. The more traditional measures are usually presented as ratios 
(e.g. relative risk). They fail, however, to impart a sense of the 
magnitude of the disparity that is achieved when using the number of 
excess deaths. The Task Force used excess deaths, therefore, as a 
methodology that more dramatically characterizes the disparities in 
health status. Because it is less familiar than other measures of 
mortality, the technique must be explained in some detail. 

The excess deaths methodology used the nonminority (White) death 
rate as a baseline from which to calculate the actual number of minority 
deaths exceeding those which would have occurred if a minority group 
experienced the same mortality rates as the White population. A death 
rate, from all causes or for a specific cause of death, is commonly 
expressed as a rate of deaths per 100,000 population. Among all 
racial/ethnic groups, death rates vary by gender, age, and cause of 
death reported. 



12 



Figure 1 shows the overall average annual death rate for White 
males and females by age group. In general, women have lower death rates 
and longer life expectancy than men at any comparable age. Death rates 
for those under age one reflect infant mortality. Relatively few deaths 
are observed among children over one year of age. A slight increase is 
observed in death rates in early adult life, largely attributed to 
external causes including homicide and accidents. A close steep rise in 
the rate of mortality is seen for adults after age 65. 

Death rates are influenced by mortality patterns related to the 
different ages at which a specific cause of death occurs. Causes of 
death confined to infancy, for example, are a major proportion of the 
deaths occurring before adulthood, whereas deaths from cancer and heart 
disease are highest in middle and later life and eventually account for 
the greatest proportion of total deaths. Consequently, the contribution 
of a particular cause of death is related to the age at which it is most 
expected. Examination of major contributions to mortality during early, 
middle, or later life will reveal differences in the causes of death 
affecting young, middle-aged, or older individuals. Gender, age, and 
cause of death differences thus became important in excess deaths 
analysis to identify the health issues of greatest concern to a 
population group. 

The excess deaths analyses of the Data Development Subcommittee 
were derived from death rates and depend on the validity and reliability 
of those statistics. Data were computed from datasets supplied by the 
National Center for Health Statistics (NCHS) annual mortality microdata 
tapes. The Task Force employed mortality data for the years 1969-71 and 
1979-81. Decennial census population data for 1970 and 1980 were used 
as the denominators to compute annualized mortality rates. National 
mortality data were available for Whites and Blacks, fewer and less 
complete data were available for Native Americans, Asian/Pacific 
Islanders, and Hispanics. 

EXCESS DEATHS METHODOLOGY 

The number of excess deaths was calculated by applying the age, 
sex, and cause-specific death rates for Whites (under age 70) to the 
comparable minority populations to derive an "expected" number of 
minority deaths. These were compared with the actual deaths that 
occurred in that minority group, computed separately by age, sex and 
cause of death. In the Task Force Report this calculation is shown as 
excess deaths = actual deaths - expected deaths. 

For those age or cause-of-death categories in which the minority 
death rate equalled the White rate, there were no "excess" deaths for 
the minority. For causes of death in which the minority rate was lower 
than the White rate, comparison with the White baseline resulted in a 
"negative" number of excess deaths. For example. Blacks under age 45 
have a lower death rate from unintentional injuries than Whites. To 

13 



prevent distortion of the number of excess deaths that actually occur 
among minorities due to the arithmetical summation of negative and 
positive numbers, negative numbers were not included in calculation of 
excess deaths (see Volume 1, page 71, Table 6, final column). Because 
of variations in death rates across the age and cause-of-death 
categories, and because the mortality profile of each minority also 
varies across these categories, excess deaths do not occur uniformly for 
each minority. Reasons for lack of excess deaths in some of these 
categories are discussed in the section on limitations of the data 
analysis. 

Although the actual death rate for any population varies according 
to age as shown in Figure 1, the excess deaths methodology uses the 
White rate for each sex, age, and cause of death as a baseline to 
evaluate minority mortality. Visualized as a graph, this comparative 
method portrays the minority death rates as they differ from the 
mortality rate of the White population for each age group. For example, 
Figure 2 shows the overall death rate for Whites as a baseline with the 
rate for Black males in each age category compared with the White rate. 
Except for ages 15-19, the mortality rate for Black males is higher than 
for White males. The disparity between the two rates illustrates the 
deaths among Black males who would not have died if their death rate 
equalled the baseline rate, mortality of White males under age 70. 
Figure 3 shows excess deaths for Black females. 

In an analogous method, the Task Force calculated the disparity 
between each minority group for whom data are available and the White 
population for age- and sex-specific overall mortality. Figures 4 and 5 
illustrate the disparity between Native American death rates and White 
death rates. As discussed in the Executive Summary , excess deaths for 
Native Americans occur principally among those under age 50. 

National data indicate that Asian/Pacific Islanders as an aggregate 
population group have an overall death rate lower than that of White 
Americans. Figures 6 and 7 show the Asian death rate as lower than the 
White baseline rate for both males and females and at all ages. This 
generalization however obscures the several small sub-groups of Asians 
with unique health needs. These issues are explored in the accompanying 
papers and subcommittee reports. 

As noted in other Task Force volumes, national data on mortality 
rates for Hispanic Americans are not uniformly or reliably available 
and excess deaths could not be calculated from this dataset. 

Following analysis of overall excess deaths the Task Force sought 
to determine the causes of death that account for the excess deaths. 
Subsequent analyses showed that more than 80% of the excess deaths 
occurring among minorities under age 70 fell into six areas. Table 1, 
reprinted from Volume I, Executive Summary , summarizes the 
contributions of the major causes of death to overall excess deaths 

14 



among the Black population. Although the actual contribution of each 
cause of death varies among the four minority groups, these six areas 
remain significant for most of the higher mortality reported among 
minorities. 

Based on examination of mortality patterns at different ages, the 
Task Force reported excess deaths only to age 70. A finding, not unique 
to this analysis, was the existence of lower mortality rates after age 
70 for Blacks and other minorities compared with Whites. This "cross- 
over" effect was seen for overall death rates and most of the major 
causes of death, including cancer, cardiovascular disease and stroke, 
cirrhosis and liver disease, and unintentional injuries. Although the 
exact cause of the "crossover" effect is not known, this phenomenon may 
reflect inaccuracies in data collection and unreliability in reporting 
age or a hardiness among those minorities who survive to later life. 
Emphasis on mortality before age 70 in the Task Force analysis avoids 
confounding of data from mortality rates of the oldest population whose 
experiences are unlike those at earlier ages. 

LIMITATIONS OF DATA ANALYSES 

Limitations of the excess deaths analysis used by the Task Force 
fall into two categories: limitations of the data base and limitations 
inherent to the excess deaths methodology itself. 

Limitations of the Data Base 

The dataset analyzed by the Task Force was severely limited by the 
lack of accurate and complete national mortality data on the Hispanic 
population. National mortality data for Hispanics were not available 
because, although the current Standard Certificate of Death adopted in 
1978 includes a race identification (White, Black, American Indian, 
etc.), it does not include an Hispanic identifier. The National Center 
for Health Statistics (NCHS) has recommended that states voluntarily add 
ethnic identifiers and has provided instructions for the Funeral 
Directors' Handbook on Death Registration and Fetal Death Reporting 
(1978 ) . The reporting of Hispanic identifiers on death certificate data 
from 1979-1981, however, was not adequate for analysis by the Task 
Force. The revised Standard Certificate of Death and the Standard Report 
of Fetal Death, proposed for implementation effective with the 1988 data 
year, however, will include an Hispanic identifier. Until that time, 
NCHS will prepare a report on Hispanic mortality for those states with 
ethnic identifiers. 

The Subcommittee on Data Development attempted to overcome the 
absence of comparable national data for Hispanics by using the country 
of birth indicator (for Cuba or Mexico) on the death certificates along 
with Cuban- or Mexican-born population data from the 1980 U.S. Census to 
suggest possible areas of highest priority for the health of Hispanics. 
This approach, however, had severe limitations in its coverage, both in 



15 



the census populations and the death certificates. For example, 
significant undercount of Hispanic mortality under one year is apparent 
when using foreign-born death certificates since the infant mortality 
rate among those born in Cuba or Mexico but living long enough to reach 
the United States is very small. Foreign-born mortality should not be 
considered as representative of the Hispanic American health status and 
cannot be acknowledged as parallel to data on other groups. 

As another attempt to obtain comparable data on Hispanics, death 
certificates for the Spanish surname and White non-Spanish surname 
populations of Texas were analyzed for the Task Force to try to identify 
disparities in health for this Hispanic population. The analysis is 
included in this volume in the paper by Bradshaw. While not national, 
these data provided useful insights and directions for recommendations 
to the Task Force and its subcommittees. 

Compared to Hispanic Americans, data on Asians are more reliably 
reported on death certificates. Death certificate data for 
Asian/Pacific Islanders are limited, however, because they reflect 
mainly the group of Asians who are the healthiest and most 
numerous — Chinese and Japanese. Asian/Pacific Islanders whose health 
needs are reported as different compared with Whites, such as Filipinos 
(high levels of hypertension) , native Hawaiians (cancer) and Southeast 
Asians (infant mortality), are few in number in national data. The Task 
Force attempted to obtain additional data on health problems of Asian 
Americans in the Pacific Basin through a review of data made available 
by the San Francisco Regional Health Administrator, Public Health 
Service (PHS/DHHS) . This analysis, conducted by the University of Hawaii 
under contract to the PHS, was seriously hampered by an absence of 
information, especially reliable vital statistics, in all jurisdictions 
in the Basin except Hawaii. 

Again, the Task Force supplemented death certificate analyses by 
commissioning consultants to analyze and compare data on health status 
of Asian Americans with that of Whites. Their papers (included in this 
volume) aided in emphasizing the needs of Asian/Pacific Islanders by 
analysis of mortality and morbidity comparisons among Chinese, Japanese, 
and Filipino Americans. 

Data from the Indian Health Service (IHS) were used to augment 
death certificate analyses and confirm the findings of the excess deaths 
analysis. Native American mortality rates compared with other population 
groups show a high death rate at early ages (before age 45) but lower 
death rates after middle age. It has been suggested that Native 
American populations have an early "crossover" effect (Figures 4 and 5) . 
IHS data emphasize health needs related to higher rates of suicide, 
unintentional injuries, and diabetes among Indian subgroups. 



16 



Limitations of the Methodology 

Limitations due to the technique of excess deaths are particularly 
important in interpreting the Task Force findings. Excess deaths are 
based on the nonminority death rate and are a comparative and a relative 
measure, dependent on both the death rates of the White population and 
the size of the minority group. The White death rate that is used as 
the base from which to estimate expected deaths, as shown in Figure 1, 
is not a constant rate, but varies according to characteristics of the 
White population. If a major cause of death among Blacks is also a 
major cause among Whites, such a cause can go unidentified with this 
comparative approach. For example, relatively higher rates of death from 
myocardial infarction among White males diminish the impact of excess 
deaths among Blacks from this cause irrespective of the actual number of 
individuals affected. 

Another limitation is that use of the number of excess deaths 
rather than death rates may underrepresent some concerns due to the 
dependence of the measure on the population size of the minority group 
and the small population base of minority subgroups. Consequently, the 
Task Force overall findings are shown most prominently for the Black 
population for which there are sufficient numbers within the population 
base to be relatively confident of priorities. Use of excess deaths is 
less appropriate to determine priorities for smaller minority groups. 

Similarly, the excess deaths technique is not useful for most 
smaller population units, for example, county or city-level analyses 
that have a small population base and fewer members of minority groups. 
Use of national data in excess deaths analysis is most appropriate to 
ensure the broadest population bases for calculation and to provide 
information for programmatic efforts that must apply to the entire 
country. 

Comparisons of minority and nonminority health status over time 
using the excess deaths techniques may be misleading unless changes in 
the White baseline rate are carefully evaluated. A change (increase or 
decrease) in death rates among the White population will modify the 
minority excess even with no change in the actual minority mortality 
rate . 

Analyses of contributors to excess deaths emphasize health needs 
that affect the greatest number of people within a population group and 
minimize needs of those with less common or rare conditions. That is, 
excess deaths combine measures of relative risk with public health 
impact as determined by the number of individuals at risk or affected by 
a condition. For example, conditions such as a sickle cell anemia may 
show great disparity in rates between the minority and nonminority 
population, but actually occur to relatively few individuals compared 
with those affected by heart disease. In planning national priorities, 
the Task Force selected areas which have a strong public health impact. 



17 



Less common health problems must still be addressed by programs at 
national or other levels. 

The technique of excess deaths as used by the Task Force was an 
appropriate tool for setting priorities relevant to policy formation, 
rather than as a methodology intended to replace other accepted 
epidemiological and investigatory techniques. Excess deaths analysis 
clearly highlighted the differences between minority and nonminority 
health status. While it allowed the Task Force initial identification 
of problem areas, it did not specify the causes of those disparities. 
Subcommittees of the Task Force were formed to do the next analytical 
step: the detailed examination of each area to determine what is known 
about the reasons for the persistence of inequities in each cause of 
death. The subcommittee reports contain findings and recommendations 
related to the causes of the disparity. 

OTHER MORTALITY MEASURES 

Other measures of mortality used by the Task Force to reinforce and 
confirm their basic findings are briefly reported here. 

Person-years of life lost has become a widely used indicator of the 
impact of mortality on public health. This measures the degree to which 
populations are affected by differences among groups in age at death. 
It is often used to compare the social, economic, or population impact 
of different causes of mortality for several groups of an entire 
population. The measure of person-years of life lost gives greater 
weight to deaths occurring earlier in life than those occurring later. 
For example, each death due to infant mortality would result in about 70 
years of life lost to society, a homicide of a 25-year old would result 
in 45 years of life lost, and death at 65 would incur 5 person-years of 
life lost. Interventions to prevent premature death may be related to 
greater societal benefit from preservation of an individual's economic 
and personal contribution. 

Seeking to examine the impact of the age of death on minority/non- 
minority differences, the Data Development Subcommittee used the measure 
of excess person-years of life lost as a modification of the more 
traditional measure. As in other Task Force data, this measure is based 
on the years lost due to excess deaths, not the total deaths for a given 
cause. Excess person-years of life lost have been calculated up to age 
70 from the Task Force analysis by multiplying the number of excess 
deaths that occurred at each age by the difference between that age and 
70. Results indicated that among Black men 914,688 years of life before 
age 70 were lost each year in excess of the person-years lost by the 
White population. Among Black females 573,159 excess person-years were 
lost annually in excess of the loss among White females. Among Native 
Americans, 34,087 excess person-years for males and 17,406 for females 
were lost annually in excess of the White population. These figures 

18 



give an indication of the personal and societal loss due to mortality 
disparities between minority and nonminority populations. 

Life expectancy also was examined by the Data Subcommittee for each 
minority group. Comparisons of life expectancy in population groups are 
a standard indicator of differentials in health. Life expectancy takes 
into account the death rates at different ages in the population at risk 
and indicates the number of years one is expected to survive from a 
given age, at birth (age 0) or after reaching a particular age. The 
person-years of life lived by each cohort are used in computing life 
expectancy for each minority. 

Table 2 (males) and Table 3 (females) show life expectancy at 
selected ages for Blacks and Whites and compare life expectancy changes 
since 1961-71. Differences over the decade of the 70 's for Whites and 
Blacks indicate gains in life expectancy for Blacks, although the 
overall disparity in life expectancy continues. Smaller differences in 
life expectancy between Whites and Blacks after age 65 demonstrate the 
cross-over effect among older Blacks. 

Relative risk of death due to a specific cause, as used by the Task 
Force, is the ratio of the minority death rate to the White rate and 
indicates the proportional risk to the minority population relative to 
the White population. Limitations of this measure as an index of 
overall health status must also be recognized. Diseases that have 
extremely low, overall mortality rates have few numbers of expected or 
excess deaths. However, relative to the White population, certain 
minority groups may show a high relative risk for those diseases. Thus 
an uncommon cause of death may appear with a high relative risk although 
it may not appear significant when considering excess deaths because its 
impact in number of persons affected may be small. Reliance on this 
measure to set national priorities may be misleading if it diverts 
attention from those causes of death which have the largest population 
impact to those which have a high relative risk but low population 
occurrence. 

Examination of differences in overall relative risk of death for 
each age group is, however, useful in suggesting targets for 
intervention. As with other measures, age-specific data available to the 
Task Force were most complete to determine mortality rate ratios for 
Blacks compared with Whites. The Task Force Data Subcommittee used 
overall ratios to examine the age categories of greatest disparity 
between Blacks and Whites. Table 4 indicates that in infancy (under age 
one) and among those age 25-54, Blacks have a death rate twice that of 
the White population and show greatest Black/White inequities. 
Combining age groups with selected causes of death, risk for major 
causes of death are presented in Table 5. The age-specific data 
reinforce the selection of the six priority areas for minority health 
activity that can affect minority health status. The cross-over effect 



19 



is shown again in these data by the ratios that are under 1.0 for older 
age categories. 

Morbidity Measures 

Although the NCHS morbidity database generally includes race and 
Hispanic identifiers in questionnaires, most data are from national 
sample surveys with proportional population representation, but in which 
minorities are not oversampled. Smaller groups, particularly Native 
Americans and Asian/Pacific Islanders, are represented in these national 
morbidity data by very few households. Consequently, population-level 
inferences of health needs are not appropriate because there are not 
enough responses to draw valid conclusions concerning the health of the 
group as a whole. The Task Force reviewed available overall morbidity 
data and urged the six subcommittees to consider morbidity in their 
individual reports. 

Four sources of morbidity data were examined by the Data 
Development Subcommittee to supplement national mortality data. First, 
several morbidity measures were taken from the National Health Interview 
Survey (NHIS) of NCHS. The NHIS is an annual survey of approximately 
40,000 households, with health information obtained on all household 
members . In order to collect data on a wider range of chronic diseases 
in recent years, households sampled for the NHIS were divided into 
subsamples and each subsample was asked about the occurrence of one of 
six different lists of chronic diseases. Because of the small numbers in 
each subsample, the Task Force found it necessary to combine several 
years of NHIS data together to make reliable estimates of the prevalence 
of many diseases. Rates for five conditions considered important by the 
Task Force are summarized in Table 6. Data from NHIS related to use of 
medical services and self-reported days of disability, Tables 7, 8, 9, 
and 10, were reviewed by the Task Force. 

The DHHS Office of Civil Rights made available to the Task Force a 
survey of hospitals under the Hill-Burton Act. This survey reports the 
annual number of admissions and emergency room visits to each hospital. 
Racial and ethnic information is available for all patients. The 
aggregate nature of the hospital data, however, precluded analyzes by 
age, sex and diagnosis, or other important characteristics. This 
database, therefore, was too limited in scope for the Data Development 
Subcommittee to derive meaningful statements concerning disparities in 
minority health or to suggest appropriate national policies for their 
alleviation. 

A relationship between health and socioeconomic status, widely 
reported in the literature, suggests that those of lower income and 
educational levels generally have higher mortality and morbidity rates. 
Nevertheless, studies reported in the research literature do not contain 
national level information on this relationship for identified diseases 
or mortality with minority status taken into account. Consequently, the 



20 



Survey of Income and Education (SIE) was explored as a possible data 
source to examine the health status of minorities and socioeconomic 
status. 

The SIE, a nationwide household survey conducted in 1976 by the 
Census Bureau, asked respondents if they had any disabling health 
problems, chosen from a list including such conditions as heart disease, 
arthritis, blindness, deafness or hearing impairment, speech problems, 
or nervous disorders. Income and education of the respondent were 
obtained. Although this is the largest national health survey of 
socioeconomic data the disease specific measures in the SIE were not 
asked in the context of a general health survey. The findings from the 
SIE thus did not indicate the cause-effect relationship of socioeconomic 
status and health. However, the SIE confirmed a general relationship 
between poor health and lower socioeconomic status for minority groups. 
A paper by Haan and Kaplan, commissioned by the Task Force also examines 
the overall relationship of socioeconomic status to health for 
minorities . 

The morbidity database of the Health Care Financing Administration 
(HCFA) was examined by the Data Subcommittee. These data, however, are 
limited to information on medicare and medicaid populations and do not 
represent health needs of an entire population. The limited data 
reported by HCFA (Table 11) suggest a higher percent of 
hypertension-related End Stage Renal Disease (ESRD) among Blacks 
compared with Whites. These ESRD data were reviewed by the Subcommittee 
on Cardiovascular and Cerebrovascular Diseases and discussed in the 
report of that subcommittee. 

The Data Development Subcommittee also examined the DHHS Health 
Data Inventory to determine the extent to which departmental databases 
include indicators of the minority status of the population covered by 
them. Departmental data were found to be primarily administrative and 
inadequate to compare health status of minority and nonminority 
Americans. For example, the Centers for Disease Control (CDC) and 
Health Care Financing Administration (HCFA) have large data collection 
activities for administration, but their information is reported in 
aggregate without race/ethnic categories. Data on Native Americans are 
provided by the Indian Health Service (IHS) and are collected only for 
those populations served by the IHS. Consequently, other departmental 
databases were not included in the Task Force analysis. 



21 



«1 

9J 












■o 
'S 



s 

05 
1/1 

S ■= 
^ I 

V Q 

'2 Cm 
.C O 

Q ^ 
^< 

^ r« 0* 







•a 



CQ 

c 
o 



5 



22 



QO 



C/5 






.•tt -2 

^^ 

o -S 
-*- « 

OS c^ 
s 

^ U 
■« — 



5 « 

u »J 

"« i 



c 
o 

3 

n. 

(£ 

o 
o 
o 
o' 
o 



a. 



^ 




^2 ^ 



< 






ajBy ajiqM UBqj jaqSiR 



ajB-^ 31!MA\. uBqi JSMoq 



23 





00 






3 




!/5 


^ 




"^ 


a 




B 






t£4 






a> 


t;« 




■«^ 








"O 




JS 


a 




^ 


'S 




© 


^ 




■^ 






9^ 


-^rf 




s; 


0^ 






Q 






«M 




'3^ 


o 




P^ 


4> 




Vi 


»i 




<u 


s 






C5 






U 






< 






a 




«J 


iXl 


ca 


n 


«8 


3 


i« 


Q^ 




S 


o 

o 
o 
o 
o 


p^ 


a; 


L-l 


m JB 




^ 






t/D 




4 


1 










ON 









ON 


ca 




f<-l 


^ 




iri 






m 


c 


»n 






00 

ON 




.a 














2 




Q 


fl 




-*-* 


•73 




03 


<u 


ON 




n: 


(N 


<U 




1 


W) 




(^1 


< 


o 



ON 


§ 


f— ^ 






»n 


£J 


1— H 


o 




u: 




^ 




f2 








•5 



ON 



<u 

C 



Q 



3lBy aJIMM uf^m JaM^IH 



9jBy ajiqAV u^MJ JSMoq 



24 



en 

2 2; 



w 
^ 



'mm 

c 



.1 



« Q 



C3 ;« 
4> U 










U-1 



ON 






ON 



ON 


S2 




cn 


C3 




r<0 


;2i 


m 




_c 


QO 

2 




J3 


S 




4—' 


(?^ 




« 






<U 


,. 




Q 


1 


ON 


^ 


X 


(N 







ON 
I 



On 



OQ 

c 
o 



•S 



o 
U 



T3 

C 

P 



c 


Pi^ 






3 


C 










•^ 


■a 


, 


D 


'R 


<u 








3 




o 






uu 


<u 




ti 


-fK 


3 






* 


to 



ajB^ sjtqAV u^ip JsqSiH 



9jB^ 9JIMAV u^Ml JSMoq 



25 









QO 

a 

Pi 

a 



I/! 

iS 



s 



Hi 

S 

U 



9^ *- 

fc, OS 

I ^ 

^ 0X1 



c 
_o 

a- 

a 

o 
o 
o 
o 
o 






.2 




■s 
■a 






2> 



sjBH 3JIIIAV UBip jgqSiH 



ajB^ SJIMM "^MJ J3MOT 



26 



"a 



Oi 



00 

Pi 

3 

Ti 

*s 



o 5 
*- « 

,> Q 

^ O 

OS en 
I/} CS 

"rt — 

u 
.o 



3 
00 






^1 



C 
O 

3 
D. 

O 

o 

O 
Q 






^ 





ON 



in 



ON 
U-1 



IT) 



On 



</-) 
^ 



ON 


« 




m 


ca 




m 


;2i 




r*) 




m 




c 


00 

ON 










x: 


oo 










cd 






Q 




ON 


m 


K 


(N 


<u 


& 


«o 


tjfl 


o 


(M 


< 


c 



2 

IT) 



ON 
I 



^ t 



'-' 




< 


u 


Q 


£:> 


■a 


a 


ili 


c 


j= 




D 


:? 


—1 



o 
o 

NO 



o 
o 



o 
o 

oo 



o 
o 



^(2 



o 
o 



o 
o 
00 

I 



T 



NO 



9jB>i sjiqAV UBqj JsqSiH 



ajB^ 9JIMM "^Ml ■iS'Vvoq 



27 



00 



S - 



S 



^1 

1^ 






2 
I 




■s 
•a 



u 

c 
o 



o 



a -S 



Q 


fv 






iM 


f^ 










^ 


y 


^ 


D 



^ 



3jB^ 3imAV u^qj -jaqSiH 



91By 311MAV UBqj J3A\oq 



28 



Table 1 



Average Annual Total and Excess Deaths in Blacks 
Selected Causes of Mortality, 
United States, 1979-1981 





Excess Deaths 
Males and Females 
Cumulative to Age 45 


Excess Deaths 
Males and Females 
Cumulative to Age 70 




Number 


Percent 


Number 


Percent 


Causes of Excess Death 










Heart Disease and Stroke 


3,312 


14.4 


18,181 


30.8 


Homicide and Accidents 


8,041 


35.1 


10,909 


18.5 


Cancer 


874 


3.8 


8,118 


13.8 


Infant Mortality 


6,178 


26.9 


6,178 


10.5 


Cirrhosis 


1,121 


4.9 


2,154 


3.7 


Diabetes 


223 


1.0 


1,850 


3.1 


Subtotal 


19,749 


86.1 


47,390 


80.4 


All Other Causes 


3,187 


13.9 


11,552 


19.6 


Total Excess Deaths 


22,936 


100.0 


58,942 


100.0 


Total Deaths, All Causes 


48.323 




138,635 




Ratio of Excess Deaths to Total Deaths 


47.4% 




42.5% 




Percent Contribution of 
Six Causes to Excess Death 


86.1% 




80.4% 





29 



u 

e 












4> 

la 
.9> 









00 



u 

13 



M ^ 

C ^ 

«« T-l 
0) QJ CK 

e2 3S 






o 



ON 

m 



m 


O 




rn 



u 






90 



0\ 

0\ 






-^ 


r- 


in 




U-5 


m 



q 

00 
NO 



in 









^ d 



01 

Of 



t^ 
^ 












in 






DC 

< 












K X 





->, 


u 


■n 




r 


















z 


OQ 


tu 




o 




Qi 




■J 




o 





30 



s 
sa 






00 



0) 

u 

S 
a> 

u 












u 

«" OS 

w e 



^ 

o 


vq 


_ 


Ov 


IT) 




od 




ON 


SO 



V3 


t~ 


Qi) 


ON 


"3 




S 


S 


01 


NC 


tb 


OS 



CO 
OS 



OS 

t-- 

OS 



^ 


u-i 


<^ 


r~ 




d 


ri 


r^ 



so 


\q 


r- 


(N 








r^ 



01 



(N 


r-- 


^ 


r- 


00 






00 



in 






so 



01 

DC 
< 









SO 



D 


i; 


u 


T3 




C 


CO 


cd 


c 




o 


.^ 




u 


a 


cd 


z 


00 


ui 




u 




as 




D 




O 





31 






% 








V 






♦j 












43 






^ 




c« 






.:« 




,X3 




u 

PS 




-tiri 




2 




•P4 




pa 




-o 






OJ 






h 






08 






CU 






B 




0) 


o 




w 


U 




s 


IX> 






^ 






u 






a 












s 






u 






o 


^H 




Sm 


00 




>%d\ 




■*rf 


t- 




mm 


o\ 




a 


^H 




■*•* 






u 


»^ 




© 


.s 




S 


C8 




"S 








© 










s 


(/3 






s 

C8 






U 




"3 


-- 


S*. 


c^ -a; 




-< 



o 

ON 



O — ' 









ON 






NO 

o 



OO 
1^ 



r^ IT) 



CM 

OO 



>n o 

O — ' 



NO 






r4 






NO — ' 



NO 
NO 






o 



C50 












< -- 



"3 


c 
D 


■* 


2 






5 

lA 


in 

t 






OO 

in 


+ 
in 


^^ 


lO 


— ' 


(N 


r<i 


T^ 


>n 


v£> 


r~ 


00 



(/3 o 

g J 

1 = 

-o J3 

« c 

.a o 



Q H 
lii 

o 
o 

00 



32 



•- On 
la *~ 



cs 



a> 



S S3 
o a> 

n OX) 



© 



C3 









DO 



+ 
IT) 



90 
I 



I 

so 






1/1 

I 
1/5 



5 

rr, 

















8 5v 


m O 

OO 00 




d d 


-^ d 


d d 


d d 


fN 00 

o\ o 


— o 


^ IT) 


■* r<-) 



SD — 
O U-) 






\0 OO 






fN O 



VO o 
00 r- 






tN OS 



en so 

so (N 



OS 00 

00 m 



On -^ 






m — 

O (N 



o — 



00 ^ 



SO OS 

o\ o 



OS 00 



00 — ' 



OS (N 



>r) o 

00 lo 



O (N 



OS OS 



r- OO 



so O 
O so 






t^ O 

so r<^ 






SO <r) 





^^ 


t^ OS 


o o 
1/-1 q 


en — 1 


n 


— ■ <N 


d d 


tN — ^ 


^^ ^^ 


* 


OS 

o o 


■* 00 

OS o 


o "n 


in so 



■* IT) 

— o 



Q 
> 



_u 


c E 




^ 


CO 

> 


aj 


c/: 


<u 


JJ 1 


c w 






O u 






13 t 


2 PU 


2 






J: u 
2 tu 




Is. tu 




Sz 






UQ 




< 





33 



en 
9i 



^^ 





*^ 












-o ^ 






a ». 






b. u 


^ 




CQ a> 


u 




a-o 


_2 




i s 


S 




so 






Vi "O 






"^ c 






^ « 






PQ ox 






u < 






a >. 






^^ 






.tS -G 






>— a -4^ 






CS C3 






t* a> 






So 






5! c^ 






O 4, 




3 


^ 3 






P^ U 




S 


0) t3 






>■ QJ 




IT) 

01 






Z 
^ 










+ 
90 



00 












3 

I 

V) 



I 



I 



I 



I 



I 



oi n 

C >" 

S-4 



d -^ 



TJ- 00 



00 U-1 



vo m 






T) 00 

00 oo 



lO o 



O \D 
O VO 



o o 



o 
o 



ii -H lu 



.2 S 



03 





(U 




> 






JJ 


J 






OJ 


o 


E 


s 


■u 


o 


tu 


-C 




U 



00 m 
o" d 



d d 



OS o\ 



o o 






<N 



00 00 



oo \o 



n o 
m IT) 



o o 



Ji 2 
2 tu 



T3 -O 



VO —I 



00 ro 















r-- o 

\o o 



^ OS 



(N O 
(N (N 



O vo 
SO (N 



CM r- 



E 
o 



<u 2 

^ ID 



^ I 



d d 



ro tN 

d d 



■<a- tN 

d d 



■* tN 

d d 



so o 
d d 



>0 Tt 

OO in 
d d 



t^ o 

d d 



tN O 

d d 



I I 



I I 



X 



3 
I/O 



<u 2 



o 

D 
O 



34 



CA 
V 




■«^ 








'.S 




^ 




JS 




•^^ 








> 




-a 


S 


<u 


op 




00 




ON 


© 


r\ 


u 


U 




4> 


1/1 


■o 


,:^ 


s 


w 


0^ 


.2 


O 


S 


■u 


b 


s 


«2 


a 


>-> 


DX) 


'-S 


< 


ifi 


>^ 


ui 


Ji 


© 


C 
© 


© 


5 




■« 


^ 
(ff 


s 
© 


5 u 


O) 


■o 


> 


V 


Table 6 

Relati 


13 

C/5 



s 



»8 

3 
O 
> 
u 
a 

Z 



2i 



V 
b 








"« 


Vi 


E 




V 

b 






<: 


_4j 








"« 














E 

■2 


w 
V 




« 


"w 


© 


>^ 


E 




C/3 


« 

u 


>i 


(I. 


al 


u 




cs 


o 




V 


■*-» 




t« 


CS 








(A 


Q 


3 

u 


"5 










c 


"a 




© 


E 




'« 


u 




B 


tb 




■*-» 






u 






a> 






a. 


en 




>^, 


U 




x 


W 



on 

-a: 



o 






00 






in 



O 






q 









<n 



00 






oo 
oo 









(SI 












o 






OO 






<n 



5 

1 
(N 



CS 






d 



ON 



s 



o 






(M 

00 






(si 



I 



OO 



O 



oo 



O 



On 



in 



IT) 



(N 
(N 



in 

oo 



(N 



o 


c<-i 


(N 


ri-i 


fsi 


rj 




CO 

m 


C3V 
en 



+ 

o 






3 

on 



U 

D 
O 



35 



s 






2 -o 
.22 C 

o 5 

c« © 
O fS 



C/3 



- ^ 

;j Pi 



c 



c 
■« aS 3 



o ■- 



B 
3 

u 



5 = 
£2 



s « 
^ E 



B 
V CQ 

> .a 



u 






Tt — . 






<3\ 



O 






00 












o 



00 









^ 
t^ 






oo 



— Tt 






s 









OO 



o 






00 



tN 









o 









ON 















as 



01 

em 



< 






5 S 









00 

vo 



d 



(N 



oo 






oo 



0\ 



00 



cn >— 



00 



O — . 



O 



oo 






m — • 



r~- 00 



2 tu 



ca 



E 

D, 

O 



t 
3 
00 



W 

u 

a: 

D 
O 



36 



u 

O ^ 
C ■== 
=3 S 

C OD 









VI © 




>^ »2 




« o 




« C 




>% © 




** '"^ 




•S '^^ 




S C8 




•P* ^^ 




-A 5 




a a 




t/! © 


00 


s^. 


^ 


-o c/5 


z 


4> 


H 


n ;^ 



b 
V 



s 



B 

n 

3 

u 



£2 



c n 

» I 





B 


a> 


« 


_> 


u 


*-S 


'C 


n 


V 


Z 


E 




■< 






Ov 


^ 


oo 


q 


* 


r^ 


vd 


vO 


"* 


— " 



(N 


fN 


<N 


r-; 


VO 


od 


MD 


VO 


00 


r-^ 



-* 

(^ 



__ _- 0^ i^ 



— 1^ 

VO ^ 



VO 






oo 
d 






00 



00 



ON 






VO 






q 






— . — r- 
iri iri r^ 



CTv 



it 

OD 
< 



u 

OB 

< 




T ■ 


s 


+ 


"-■ 


c 


r^ 


m 


VO 


< 


"^D 




^ 


VO 



en 
VO 






On 

od 



ov r-^ 



q --; 
1/^ p~ 



vo q 
d --^ 



q 
od 



d 



C/3 



2 u. 



1/^ 


vO 


S 


in 


r-^ 


S X 



§2 



u 

OS 

O 



37 






u, 




ca 




<u 




>H 




-<-< 




(» 




(R 




&H 


^^^ 




GO 


0£ 


>^ 


G 


£! 






'Z 


X5 


3 


s 


fi 


CO 


S 

o 


DO 


-^^ 


< 


C3 


>> 




^ 


s 




2 




■4^ 


3 
O 




Ln 


*> 


o 


■fcd 






c 
o 








s 
« 


¥ 


3 

a 


■«irf 


o 








C« 



O -2 



B 
3 

U 



£2 



1 


B 


B 


n 


ca 


_o 


u 






*C 


'S 


OJ 




E 


;* 


< 



a 

> .a 

IS V 

Z S 









* 

o 






? 



o 


lO 


^ 


r<^ 


_ 


r-~' 


ro' 


ON 


rn 


(N 



oo 


CN 


a\ 


r- 


■o 


■*' 


^3 


ro 


ON 


(N 






'~' 


fN 


lO 



"1 


o 


■^ 


lo 


^ 


ON 


m' 


r- 




cm" 



'd- 


00 


~ 


00 


q 


On 


iri 


od 


ON 


ON 

in 



rn 


r- 


m 


rg 


fN 


"O 


rn 


O 




>n 



r- 


o 


U-) 


^-H 


(^ 


2; 


'^ 


od 


fN 





01 
DC 
< 



< 



5 - 



? s 



* 

o 

On 



O 
O 



in 

od 



ON 

in 



ON 

«n 



en 
0\ 



NO 

On 



NO 
ON 



ON 



in in 



in 



in 

NO 







■t3 




!i E 


ON 


^ 


S 




^ < 
^1 


^' 


2 


I 




1-H 






^ 










■a u 






aj 


X 


=5 t 






c 


>. 


S <S 






u 










U 

c 
o 




l» eg 










O 1^ 






ca 


•a 


c u 




-S 


z 


CO 


^ 1 


(L> 


"ca 


ui 




•a .2? 


cd 


E 


O 




u -= 


S 




D 




.2? £ 






38 



u 




a 




0) 




>^ 




'*' 




tn 




a 




Ph 




9i 




A 




•^ 




D£) 




S3 




• w* 




u 




9 




Q 




e 




o 




Vi 




u 




<u 




Oh 




4> 




t 


C/2 


"d- 


-o 


a 




» 


0) 


_>;> 

■*>< 








bi 




s 


!/3 


U 


a 




s 


b 


o 


-?^ 


u 


P^ 


O 


■4^ 




VI 


s 


O 


_o 


hJ 


%-d 




i2 


V3 




>-> 


3 


c« 


a 


fi 


© 


® K^ 


Pn 


'H ^ 




1 o 


c/5 


1^ 


P 






s 

A 

3 

u 



2 = 

U U 

(£2 



.a u 

si 



c 

> .a 

Z S 

< 









r^ 


vO 


* 
w-1 


v£> 


>/i 


«-i 


I/S 


vd 






00 






Z 






oo 






O 



o 

M2 






< 
Z 



< 

z 



oo 



00 



r-1 oo 






< 

z 



00 



1/^ 



00 



O 



+ 



at 

< 



C/5 


trt 








00 


P3 








< 




^ 


s 


+ 


"•"] 


r-. 


r-- 


m 


IT) 


< 






^ 


VD 



0^ 






q 






oo 
iri 



00 






>/-i 
^ 



< 
Z 



< 

z 



t-- oo 



^ — ■ 

rt in 









CO rt t— 

oj !> o 



z m 

ui 
u 

D 

o 



39 



<» 



J3 



<y 




C 




C 




0£ 


a 




• !■ 




Q 




T3 




C 




Rt 




4> 




U 




ea 




0< 




>- 




.a 




y_^ 




fi 




Q^ 




C/3 




(ij 




w- 




O) 


ri 




00 
OS 

1-H 






5 






-4iri 


« 


s 


c 








SI 


-^^ 


D£ 


c 


CQ 


w 


-4^ 




C/3 


a 




1/1 


■a 


s 


S 


« 


S 




o 


■a 


0^ 


c 


u 


C8 


B 




_aj . 


M 


13 


33 
>> 


a; 


« 


i:5 









CI 
0\ 



OS in 

o o 

^ tNl 



o 



o 



m in 



O 



00 
ON 



O t^ Tt 

ni 00 00 

— VO 00 

IN -^ ~ 



fe? 








On 


00 


00 


m 


^—^ 


vo 


On 


CO 










CM 








<N 








00 











6? 










NO 










NO 








01 


^^ — 


ON 


r- 


oo 


^ 




On 


-* 




s: 




-* 


rn 


rr) 


^ 


r~- 


^ 


•^ 




NO 










r^ 











o 

c 


_« 








M 


C/3 








re 


O 










u 






c 


5 








ID 


>v 


0) 


c 






E 


2 

CO 


o 
c 

c 






00 


o. 


& 


E 

3 


2 




5 



H 



c 
u 
o 

u 
a. 



3 

2 



o 


so 




c 






oa 


CO 




w 


O 






u 




Q 


Q, 




>> 


ID 


c 


2 


o 

CO 


,E 


CO 


C 






OJ 






t; 


00 

c 


a. 


o. 


u 


>. 


2 


K 


ca 






u 


^ 


r^j 



X) 

CO 



CO S 











a o 


.-—^ 








« c 


fe5 








c^S 


^ 


O 


00 


NO 


Q -o 




(N 


t^ 


NO 


Cn) 


00 




Tj- 


W J«i 


o 








c » 


en 
















O CQ 





£ 


<r 


-U 






nn 


« 


c 


H 






f > 




c 

C 


c 





3 


1) 




X 


a 


w 




u 




Bi: 




D 




o 




m 





40 



Excess and Deficit 
Mortality Due to Selected 
Causes of Death and Their 
Contribution to Differences 
in Life Expectancy of 
Spanish-Sumamed and 
other White Males— 1970 
and 1980 



Benjamin S. Bradshaw, Ph.D. 

Associate Professor of Demography 

The University of Texas Health Science Center at Houston 

School of Public Health 

Houston, Texas 

W. Parker Frisbie 

The University of Texas at Austin 
Population Research Center 
Austin, Texas 

Clayton W. Eifler 

The University of Texas Health Science Center at Houston 
School of Public Health 
Houston, Texas 



A portion of this report wds presented us "Contributions of Selected Causes of 
Death to Differences in Life Expectancy of Spanish Surnained and Other White 
Males — 1970 and 1980," at the annual meeting of the Southern Regional 
Demographic Group, Orlando, October 17-19, 19y4. 

The authors wish to thank the Bureau of Vital Statistics, Texas Department of 
Health, for the mortality aata on which this report is based. 



42 



EXCESS AND DEFICIT MORTALITY DUE TO SELECTED CAUSES OF DEATH AND 
THEIK CONTRIBUTION TO DIFFERENCES IN LIFE EXPECTANCY OF 
SPANISH SURNAMED AND OTHER WHITE MALES— 1970 AND 1980 

Abstract 

In comparing populations, it is common that each will have advantages 
relative to the others with respect to certain causes of death. However, for 
some groups, particularly Hispanic populations in the United States, little is 
known about the disadvantages or advantages of their mortality patterns with 
respect to the majority non-Hispanic white population. In this analysis, 
patterns of "excess" mortality by cause of death in the male Spanish surnamed 
population of Texas are contrasted with the patterns ooserved among other 
white males in 1970 and 1980, and the impact of changes in cause-specific 
death rates during the 1970s on life expectancy is examined. 

The results demonstrate that, while there is only a small amount of 
excess mortality overall airiong Spanish surnaaed males, there are important 
causes of death, such as diabetes mellitus, pneumonia and influenza, motor 
venicle acciaents and nomicide, in which there is a large excess. If Spanish 
surname males were to experience the age-specific death rates of other white 
males, they would suffer many fewer deaths due to these causes — for example, 
about bl percent fewer Iran diabetes and 7^ percent fewer from homicide — and 
their life expectancy would increase about ^.'j, years. On the other hand, 
Spanish surnaiie males enjoy a substantial advantage with respect to some other 
major causes of death, notably malignant neoplasms and circulatory diseases. 
The fact that Spanish surname males are disadvantaged mainly fom deaths due to 
conditions which may be successfully managed (as diabetes) or to events which 
may be prevented (as motor vehicle acciaents) or avoided (as homicides) 
suggests their life expectancy might be fairly easily improved. 



43 



EXCESS AND DEFICIT MORTALITY DUE TO SELECTED CAUSES OF DEATH AND 
THEIR CONTRIBUTION TO DIFFERENCES IN LIFE EXPECTANCY OF 
SPANISH SURNAMED AND OTHER WHITE MALES— 1970 AND 1980* 

In comparing two populations it is common that each will be found to 
have advantages relative to the other with respect to some causes of death. 
If the advantages are large, the implication is that if mortality could be 
reduced in the disadvantaged population to the level in the other, then many 
lives might be saved (or deaths might be postponed). Thus, in developing 
policies for the improvement of the public health, it is useful to examine the 
disadvantages in mortality, usually expressed in terms of "excess" mortality, 
which some groups may suffer. It may be that some or all of the disadvantages 
can be removed. 

Racial subgroups in the United States vary considerably in mortality 
levels. These variations are well documented, and for the two largest 
subgroups, the white and black populations, mortality statistics are routinely 
available. The general outlines of the mortality disadvantages of the black 
population compared with the white are commonly known, ana because of the 
ready availability of information, easily demonstrable. For certain other 
groups, particularly minority subgroups of the white population such as the 
Mexican American population, much less is known about the disadvantages or 
advantages of their mortality patterns with respect to the non-Hispanic white 
population. In this analysis, the mortality of the Mexican American male 
population of Texas is described, and their mortality is compared with that of 
the other white male population. 

Disadvantages in mortality by cause can be analyzed by several 
different means. The most straightforward is by the simple comparison of age 
and cause-specific death rates. These can be summarized by different 
standardization techniques if the pattern of age-specific rates suggests that 
this would be reasonable. Another procedure is to obtain the number of deaths 
due to some cause which would occur in a population if it experienced the 
rates of another population. This produces an estimate of excess (or perhaps 
deficit) mortality. Another very useful method is to demonstrate the impact 
that equalizing age and cause specific mortality would have on the life 
expectancy of one population with respect to another. 

In this report measures based on all these methods are presented for 
Spanish surname and other white males in Texas in 1970 and 1980. First, age 
and cause-specific death rates and estimates of excess and deficit mortality 
are analyzed for the former group under the assumption that they experiencea 
the rates for other white males. We have incluaed what we refer to as 
"deficit" mortality, because, as will be shown, the Spanish surname male 
population enjoys certain very important advantages in some causes of death. 
Advantages and disadvantages in terms of implications for life expectancy are 
illustrated in the second part of the reported results. 



*Throughout this report the terms "Spanish surnaned" and "Mexican American" 
will be used interchangeably. It is recognized that the populations referred 
to by these terms are not completely identical. Problems of identification of 
the Mexican American population are discussed in the section on "Data and 
Methods." 



45 



EARLIER STUDIES OF MEXICAN AMERICAN MORTALITY 

Over the past 10 to 15 years, Mexican Americans have been the 
subject of a growing body of research reflecting an increased interest by 
social scientists in general and demographers in particular. Nevertheless, 
one very important demographic dimension of the Mexican American population, 
viz., their mortality experience, has been relatively neglected. As Schoen 
and Nelson (1981) point out, most mortality research focused on Mexican 
Americans has been limited to studies in one or two cities, to a specific age 
group (infants), or to a specific cause (notably cancer). Prior to the Schoen 
and Nelson research using circa 1970 data for California, "only one study of 
Mexican American mortality (had) been done which used a statewide population 
and included all ages and causes of death" (Schoen and Nelson, 1981:260) — a 
work by Bradshaw and Fonner (1978) who dealt with Mexicdn Americans in Texas 
and also made use of circa 1970 data. Other recent analyses of Mexican 
American mortality in Texas have appeared, but the aim has been primarily to 
assess the appropriateness of the various indicators of Hispanic ethnicity 
available in the vital statistics and census data and the consistency of 
mortality estimates based on them (Bradshaw and Frisbie, 1983; Gillespie et 
al., 1983; Sullivan et al., 1983; 1984a, 1984b). 

The concentration of mortality research on Mexican Americans in 
California and, especially, Texas is useful, given the fact that the vast 
majority of Mexican Americans reside in these two states (over 73 percent in 
1980; see Davis et al., 1983:13). In addition, there are no national data on 
Mexican American mortality, though an increasing number of states now include 
a question on Hispanic origin of decedents on death certificates (National 
Center for Health Statistics, 1983). The present research continues the focus 
oti Texas and compares the mortality of Spanish surname and other white 
(non-Hispanic white) males. 

In the 1969-71 period, in botn Texas and California, life expectancy 
at birth among Mexican American (Spanish surname) males was fairly similar to 
that among "Anglo" (other white) males, but the timing of death in the two 
groups was dissimilar (Bradshaw and Fonner, 1980; Schoen and Nelson, 1981). 
In both states there was a "crossover" pattern such that Mexican American 
males had higher death rates at young ages and lower rates at older ages 
compared to Anglo males. The crossover was the result of lower death rates 
among Mexican Americans due to circulatory diseases and neoplasms — diseases 
which are much more canmon at older ages — and higher rates due to certain 
other causes, such as accidents and hanicide, which are concentrated at 
younger ages. These results are consonant with earlier studies of Mexican 
American mortality which employed data for local areas or specific causes of 
death (cf. Ellis, 1959 and 1962; Kautz, et al., 1981; Lee, et al., 1976; 
Roberts and Askew, 1972; Stern and Gaskill, 1978). Significantly, the results 
reported for Mexican American males are quite similar in pattern to those for 
Puerto Rican males. Rosenwaike, e.g., found "markedly low mortality ratios" 
fran heart disease and malignant neoplasms among Puerto Ricans in New York 
City, but high ratios due to homicide (1983:378, 400). This suggests that 
these characteristics of mortality, at least with respect to cancer and heart 
disease, may be common among other groups of Hispanic American males. 

DATA AND METHOD 

Data sources and quality. - Data for this research were obtained fran 
death certificates of decedents resident in Texas in 1969-71 and 1979-81 made 

46 



available by the Bureau of Vital Statistics of the Texas Department of Health. 
These records provide the numerators necessary for computing death rates. 
Data for the denominators are taken from the 1970 and 1980 U.S. Census 
enumerations. 

Interest here is in comparing the mortality experience of the white 
Spanish surname population with the mortality of other whites in Texas. 
Spanish surname, in essence, identifies the Mexican American populatioiK It 
can be argued that self-reported ethnicity is a more appropriate means of 
defining Mexican Americafis, since some misciassification arises with the use 
of surname due to: 1) exogamy of females who take the surnames of their 
husbands; 2) failure to report surname on death certificates; and 3) the fact 
that not all Spanish surnamed individuals are of Mexican origin. While 
potentially troublesome in other contexts, none of these problems is of more 
than negligible magnitude in the present analysis. Our research focuses only 
on males so that the issue of exogamy is entirely moot. Second, only a very 
minor fraction of the death certificates do not contain a surname (see 
Sullivan et al., 1984b:5). Finally, Bradshaw and Frisbie (1983) demonstrate 
an extremely high degree of congruence (in the 1980 data) between Spanish 
surname decedeJits in Texas and Mexican origin as reported by survivors. This 
close correspondence is to be expected based on census records which show that 
the vast majority of Spanish surnamed persons in Texas are of Mexican origin 
and that the vast majority of Mexican origin persons in that state have a 
Spanish surname. 

Even more to the point, reports of ethnicity are not available from 
vital statistics prior to 1980 (and even in that year some death certificates 
lacking ethnic origin items were still in use; cf. Sullivan et al., 1984b). 
Thus, it is impossible to make comparisons of mortality over time based on any 
identifier other than Spanish surnaine. 

Coding of surnames on the death records for 1970 was done manually 
according to the 1970 Census list of Spanish surnames, and that for 1980 was 
by means of a canputerized nane classification system. The codes were 
assigned as part of the routine processing of death records in the Bureau of 
Vital statistics. The quality of manual coding was found to be extremely good 
(see Bradshaw and Fonner , 1978: 267). Empirical assessment of the computer 
coding procedure yields results that are highly consistent with those obtained 
by matching surnames with the 1980 census list of surnames (Bradshaw and 
Frisbie, 1983). Thus, the death data and the population data for the Spanish 
surnane population around the time of the 1970 and 1980 census are 
consistently defined , so that death rates and measures based on them can be 
computed with confidence. 

The causes of death selected for analysis in this report are described 
in terms of their International Classification of Disease codes. Eighth and 
Ninth Revisions, in Appendix A. The cause categories are standard ones, 
except for "major circulatory diseases." For our purposes this group of 
causes includes only ischemic heart diseases (acute and chronic) and 
cerebrovascular disease, and excludes other less important circulatory 
conditions. The unavailability of comparably tabulated data for 1981 required 
this adaptation. Also, the category referred to as "homicide" in the text and 
tables includes a very small but undetermined number of deaths due to legal 
intervention. 

Methods of analysis. - So-called excess and deficit mortality were 
canputed by multiplying the difference between the age and cause specific 
death rates for the two groups of males in 1969-71 and 1979-81 by the 1980 
population age distribution of Spanish surname males, which served as a 

47 



RESULTS 

Excess and Deficit Mortality. - The crossover in age-specific death 
rates that was alluded to earlier for 1970 is present also for 1980. As may 
be seen in Table 1, the rates for 1980 for both groups of males are lower than 
for 1970. The crossover of the Spanish surname rates below those of other 
white males occurs at a somewhat later age in 1980 (45 to 49 years) than in 
1970 (35 to 39). 

At first g,lance, a crossover in age-specific rates might seem to 
suggest that summarization by means of a procedure such as direct 
stanaardization is not appropriate. However, directly standardized rates for 
ail causes of death combined are highly instructive statistics since they show 
to what extent the two populations are similar in overall mortality — i.e., to 
what degree they experience the sane mortality result — despite dissimilar 
patterns of age-specific mortality. Standaraized rates are also useful for 
summarizing mortality experience by cause of death, because the numbers of 
deaths due to specific causes are not evenly distributed by age but instead 
are concentrated at certain ages. Thus, a crossover, considering all causes 
canbined, should be traced to dissimilar rates due to causes whose impact is 
greatest only at certain ages. For exaiiple, cancers and circulatory diseases 
are most important at middle age and beyond while homicide takes its greatest 
toll at younger ages, particularly ages 15 to 39. 

Table 2 shows airectly standardized death rates from all causes and 
selected causes for 1970 and 1980. The overall rate for both groups of males 
declined during the decade, and their relative relationship remained about the 
same. On examining the rates for the various causes of death, it is 
immediately apparent that Spanish surname and other white males differ 
markedly in their mortality patterns by cause: The death rates of non-Spanish 
surname white males due to cancers and major circulatory diseases are 
substantially higher. This pattern is observable for both I970 and 198O, but 
the gap between the two groups widened with respect to cancers and narrowed 
with respect to circulatory diseases because the rates of change by cause 
differed in the two populations. Lung cancer death rates increased more 
slowly among Spanish surname males, and rates for other cancer declined more 
rapidly. At the same time, major ischemic heart disease mortality declined 
more rapidly among non-Spanish surname white males between 197O and 1980, but 
their death rate from this cause was still about 13 percent higher at the end 
of tne decade. On the other hand, Spanish surname males experienced much 
higher rates due to diabetes, pneunonia and influenza, external causes, and 
"all other" causes (a residual category which includes infantile causes of 
death as well as a variety of otners) . During the 1970s there was some 
narrowing of the difference in rates for some causes, including pneumonia and 
influenza, motor vehicle accidents, and "all other" causes, but there was 
significant widening of the difference for others, particularly homicide. 

The implications of the mortality patterns of Spanish surname and 
other white males, as indicated by the standardized rates, for an analysis of 
"excess" mortality in the former group relative to the latter are substantial. 
Clearly, Spanish surname males have some mortality disadvantages, but equally 
clearly they have important mortality advantages, and these advantages and 
disadvantages tend to offset one another , so that there is little net 
difference in mortality of the two groups as measured by the directly 
standardizea death rate. Therefore, in the following paragraphs we present a 
discussion of botn "excess" and "deficit" mortality. 

Table 3 shows total, excess, and deficit deaths that would occur in 

48 



TABLE 1 



AGE-SPECIFIC DEATH RATES FOR SPANISH SURNAME AND OTHER WHITE 
MALES-TEXAS! 1969-71 l 1979-81 
(Rates per 100,000 population.) 

1969-71 1979-81 
I I 

SPANISH OTHER RATIO SPANISH OTHER RATIO 
A6E SURNAME WHITE SURNAME WHITE 



UNDER 1 


2357.21 


2150.55 


1.10 


1559.86 


1124.55 


1.39 


1-4 


136.37 


92.35 


1.48 


82.73 


71.89 


1.15 


5-9 


56.20 


50.26 


1.12 


35.09 


36.78 


0.95 


10-14 


49.99 


55.49 


0.90 


36.24 


39.72 


0.91 


15-19 


196.32 


155.74 


1.26 


205.92 


160.37 


1.28 


20-24 


341.31 


187.05 


1.82 


319.73 


206.67 


1.55 


25-29 


316.10 


166.15 


1.90 


289.87 


192,47 


1.51 


30-34 


275.14 


193.01 


1.43 


266.42 


187.97 


1.42 


35-39 


321.05 


259.63 


1.24 


296.56 


219.58 


1.35 


40-44 


436.77 


444.10 


0.98 


395.11 


330.65 


1.19 


45-49 


675.81 


720.64 


0.94 


586.66 


535.29 


1.10 


50-54 


920.85 


1106.22 


0.83 


868.65 


888.49 


0.98 


55-59 


1506.64 


1731.75 


0.87 


1281.72 


1391.94 


0.92 


60-64 


2386.71 


2582.25 


0.92 


1924.00 


2136.00 


0.90 


65-69 


3496.46 


3819.54 


0.92 


2975.75 


3249.40 


0.92 


70-74 


5535.13 


5714.84 


0.97 


4719.10 


4930.66 


0.96 


75-79 


8362.37 


8654.13 


0.97 


6774.49 


7443.97 


0,91 


80-84 


11576.29 


12240.85 


0.95 


10513.42 


11303.30 


0.93 


85 & OVER 


16216.00 


18369.07 

=========ZS 


0.88 

:==z==z= 


16243.75 

zzzzzzzzz 


18518.95 

zzzzzzszzss 


0.88 



49 



TABLE 2 



DIRECT-ADJUSTED DEATH RATES FOR SELECTED CAUSES DF DEATH FDR SPANISH SURNAHE AND OTHER 
WHITE MALES-TEXAS! 1969-71 k 1979-81* 
(Rates per 100,000 population.) 

1969-71 1979-81 
1 1 

SPANISH OTHER X SPANISH OTHER X 

CAUSE OF DEATH SURNAME WHITE diH. SURNAME WHITE diH. 
1 1-—. 

ALL CAUSES 665.30 648.60 2.57 570.90 548.40 4.10 

ALL MALI6NANT NEOPLASMS 82,40 101.70 -18.98 77.20 103.00 -25.05 

LUNG 19.10 33.20 -42.47 20.70 37.50 -44.80 

OTHER 63.30 68.50 -7.59 56.50 65.50 -13.74 

DIABETES HELLITUS 16.20 7.20 125.00 13.40 5.30 152.83 

MAJOR CIRCULATORY DISEASES 190.20 234.70 -18.96 135.10 151.50 -10.83 

MAJOR ISCHEMIC HEART DIS. 141.30 186.00 -24.03 104.40 120.60 -13.43 

ACUTE MYOCARDIAL INFARC. 92.00 122.00 -24.59 70.00 81.00 -13.58 

CHRONIC I.H.D. 49.40 63.90 -22.69 34.30 39.60 -13.38 

CEREBROVASCULAR DISEASE 48.80 48.80 0.00 30.70 30.90 -0.65 

PNEUMONIA & INFLUENZA 25.40 19.80 28.28 13.20 11.20 17.86 

ALL EXTERNAL CAUSES 143.90 104.80 37.31 143.90 103.40 39.17 

MOTOR VEHICLE ACCIDENTS 58.50 43.40 34.79 49.80 43.10 15.55 

HOMICIDE 30.40 9.50 220.00 47.40 12.30 285.37 

ALL OTHER EXTERNAL CAUSES 55.00 51.90 5.97 46.70 47.90 -2.51 

ALL OTHER CAUSES 207.30 180.30 14.98 188.30 173.90 8.28 

^Standard is 1980 Spanish surnase iale population. For ICDA definitions of causes, see 
Appendix A. 



50 



the 19ao Spanish sun^aine male population under 1970 and 19bO age and cause 
specific rates. To reiterate, excess deaths are defined as deaths which would 
not have occurred had Spanish surname males had the age-specific death rates 
of otner white males due to certain causes, specifically those causes for 
which Spanish surname males have higher death rates. Deficit deaths are 
aaditional deatns which would occur among Spanish surname males if they had 
tne age-specific death rates of other white males due to certain causes, 
namely those causes for which Spanish surname had lower death rates. Note 
that in some causes, such as cancer, from which non-Spanish surname males die 
more frequently, there are still some excess deaths among Spanish surname 
males. This is because at a few ages, death rates from these causes are 
somewhat higher among the Spanish surname males. 

Altogether, in both 1970 and 1980, if Spanish surname males had 
benefited from having death rates equal to tnose of other white males in those 
ages and causes in which tne rates of the latter group were lower, about 14 
percent fewer deaths would have occurred in the Spanish surname male 
population. The standardized death rates would be reduced by the same 
percentage, making them well below those for other white males. 

The effect of equalizing (or optimizing) Spanish surname rates with 
respect to those of other white males would have a powerful effect on deaths 
from some causes. Deaths from diabetes mellitus, for example, would be 
reduced by about 57 percent in 1970 and 61 percent in 1980. Reductions fran 
homicide mortality would be even more dramatic — almost 70 percent in 1970 and 
74 percent in 1980. Pneumonia deaths among Spanish surname males would have 
been reduced nearly a quarter in 1970, but by a lesser amount in 1980. 

Among the individual causes shown, deaths from homicide comprise the 
largest single category of excess deaths in both years. This cause accounted 
for 22 percent of all excess deaths in 1970, and 44 percent in I98O — a 
consequence of the large increase m homicide death rates in the Spanish 
surname male population . Taken togetner , the external causes of death made up 
48 percent and 57 percent of all excess deaths in 1970 and 19bO, respectively. 
Were it possible only to have avoided excess external cause mortality in the 
Spanish surnane population, the age standardized death rate from all causes 
for that group would have been reduced below that of other white males. In 
fact, elimination of excess homicide deaths alone would accomplish that 
result. 

Spanish surname males would benefit considerably if their death rates 
due to certain causes were equal those of other white males. However, they 
would suffer almost as many adaitional deaths if their death rates were to 
come to equal those due to other causes. From a policy perspective, it may be 
appealing to assume elimination only of excess mortality in a minority 
population. However, it is useful also to consider the possible mortality 
advantages that such a population may already have with respect to the 
population with which it is being compared. As a minority population becomes 
more like the majority population socially and econonically, eliminating 
excess deaths due to some causes may be accompanied by changes in mortality 
patterns that lead to increased risk of death due to other causes. 

Table 3 shows, as "deficit" deaths, the additional deaths that the 
Spanish surnane male population woula suffer if it were to experience the age 
and cause specific death rates of other white males, i.e., if its death rates 
for those causes in which it enjoys an advantage were to equal those of other 
white males. The expected number of Spanish surname male deaths would have 
increased 12 percent in I970 and 10 percent in 1980, nearly offsetting the 
excess mortality described above. Under these conditions, Spanish surname 



51 



TABLE 3 

TOTAL, "EXCESS," AND "DEFICIT" DEATHS, AND IMPLIED NET DIFFERENCE IN THE NUMBERS OF DEATHS EXPECTED IN THE 1980 

SPANISH SURNAME MALE POPULATION (SUMMED OVER ALL AGES) --TEXAS; 1969-71 & 1979-81 

UNDER 1970 RATES 

TOTAL "EXCESS DEATHS" "DEFICIT DEATHS" NET DIFFERENCE 

CAUSE OF DEATH EXPECTED 1 i 

DEATHS NUMBER PERCENT NUMBER PERCENT NUMBER PERCENT 

fiLL CAUSES 8896 -1264 -14.21 1041 11.70 -223 -2.51 

ALL MALIGNANT NEOPLASMS 1102 -22 -2.02 280 25.42 258 23.40 

Li'H^, 255 -9 -3.61 198 77.60 189 74^00 

OTHER 847 -13 -1.54 82 9.71 69 8.17 

DIABETES MELLITUS 216 -122 -56.53 2 0.98 -120 -55 55 

MAJOR CIRCULATORY DISEASES 2543 -77 -3.04 673 26.47 596 23*43 

MAJOR ISCHEMIC HEART DIS. 1890 -3 -0.16 599 31.71 596 3l'55 

ACUTE MYOCARDIAL INFARC. 1230 -3 -0.24 404 32.87 401 32*63 

CHRONIC I.H.D. 660 -0.01 195 29.56 195 2954 

CEREBROVASCULAR DISEASE 653 -74 -11.38 74 11.31 -0 07 

PNEUMONIA i INFLUENZA 339 -80 -23.61 6 1.77 -74 -21*84 

ALL EXTERNAL CAUSES 1924 -601 -31.25 79 4.11 -522 -27*14 

MOTOR VEHICLE ACCIDENTS 782 -209 -26.77 8 1.02 -201 -25! 75 

HOMICIDE 406 -282 -69.37 2 0.53 -279 -68*84 

ALL OTHER EXTERNAL CAUSES 736 -110 -14.97 69 9 35 -41 -5 62 

ALL OTHER CAUSES 2772 -361 -13.03 1 o!o2 -361 -13.*01 

UNDER 1980 RATES 



TOTAL "EXCESS DEATHS" "DEFICIT DEATHS" NET DIFFERENCE 
EXPECTED —I I 

DEATHS NUMBER PERCENT NUMBER PERCENT NUMBER PERCENT 



7634 -1071 



-14.02 



769 10.07 



-302 



-3.95 



ALL CAUSES 

ALL MALIGNANT NEOPLASMS 

LUNG 

OTHER 
DIABETES MELLITUS 
MAJOR CIRCULATORY DISEASES 

MAJOR ISCHEMIC HEART DIS. 
ACUTE MYOCARDIAL INFARC. 
CHRONIC I.H.D. 

CEREBROVASCULAR DISEASE 
PNEUMONIA & INFLUENZA 
ALL EXTERNAL CAUSES 

MOTOR VEHICLE ACCIDENTS 

HOMICIDE 

ALL OTHER EXTERNAL CAUSES 
ALL OTHER CAUSES 

NQTE-"E)icess deaths" are deaths that would have been avoided or postponed i^s7a7uh™u7n^".7.^7eTh^7hTt"e^^^^^^^ 

le !ld^'^; 'S ^M I I °'^'' "^'^' "'^"" ""'^^^'^ ''"'^^" *" ^"^'^'""^l ""ths that Spanish surna.e 
nal 5 would have ha if they had experienced the age and cause specific rates of other white .ales. "Net difference" 

;!t 'n;"nl? J^ -ie^^ase in deaths in the 1980 Spanish surname «ale population i.plied by application of death 

L !! "■ ^°'?^ ''""''' ''''''' '' '^' ""'^^^ ''^''''' "y ^PPlyi"9 the ag and cause specific 
death rates of Spanish surnaae sales. For 1980 these equal the observed deaths. ' ' " ** '' 



1032 


-7 


-0.67 


353 


34.25 


347 


33.59 


277 


-1 


-0.26 


226 


81.71 


226 


81.45 


755 


-6 


-0.81 


127 


16,84 


121 


16.03 


179 


-109 


-61.16 


1 


0.57 


-108 


-60.58 


1805 


-61 


-3.39 


282 


15.60 


220 


12.21 


1395 


-4 


-0.31 


222 


15.88 


217 


15.57 


936 


-1 


-0.15 


147 


15.76 


146 


15.61 


459 


-3 


-0.63 


74 


16.13 


71 


15.50 


410 


-57 


-13.86 


60 


14.64 


3 


0.79 


176 


-28 


-15.79 


1 


0.84 


-26 


-14.95 


1924 


-609 


-31.65 


68 


3.55 


-541 


-28.09 


666 


-98 


-14.76 


9 


1.40 


-89 


-13.36 


634 


-469 


-73.93 





0.06 


-468 


-73.87 


624 


-42 


-6.71 


59 


9.40 


17 


2.69 


2518 


-256 


-10.18 


63 


2.51 


-193 


-7.68 



52 



TABLE 4 

LIFE EXPECTANCY BY ABE FDR SPANISH SURNAME AND OTHER NHITE 
MALES-TEXAS: 1969-71 k 1979-Bl 

1969-71 1979-Bl 
I— - I — 

SPANISH OTHER DIF- SPANISH OTHER DIF- 
A6E SURNAME WHITE FERENCE SURNAME WHITE FERENCE 



0-1 


67.67 


68.09 


0.41 


70.20 


70.63 


0.43 


1-5 


68.27 


68.55 


0.2B 


70.13 


70.44 


0.31 


5-10 


64.63 


64.80 


0.17 


66.35 


66.64 


0.29 


10-15 


59.81 


59.95 


0.15 


61.47 


61.76 


0,29 


15-20 


54.95 


55.11 


0.16 


56.57 


56.87 


0.30 


20-25 


50.46 


50.52 


0.06 


52.13 


52.31 


0.18 


25-30 


46.29 


45.97 


-0.31 


47.92 


47.83 


-0,10 


30-35 


41.98 


41.33 


-0.65 


43.59 


43.26 


-0.32 


35-40 


37.53 


36.71 


-0.B2 


39.13 


38.65 


-0.49 


40-45 


33.09 


32.15 


-0.94 


34.68 


34.04 


-0.64 


45-50 


28.76 


27,81 


-0.95 


30.32 


29.57 


-0.75 


50-55 


24.66 


23.74 


-0.92 


26.14 


25.30 


-0.85 


55-60 


20.70 


19.94 


-0.76 


22.18 


21.32 


-0.86 


60-65 


17.11 


16.50 


-0.61 


18.48 


17.67 


-O.Bl 


65-70 


13.95 


13.42 


-0.53 


15. OB 


14.37 


-0.71 


70-75 


11.13 


10.72 


-0.42 


12.09 


11.46 


-0.63 


75-80 


8.90 


8.44 


-0.45 


9.65 


B.97 


-0.68 


B0-B5 


7.27 


6.72 


-0.55 


7.55 


6.91 


-0.64 


85 I OVER 


6.17 


5.44 


-0.73 


6.16 


5.40 


-0.76 



53 



deaths due to malignant neoplasms would nave been greater by 25 percent in 
1970 and 34 percent in 1980. Lung cancer aeaths would have been increased by 
more than 75 percent in both years. Under 1970 rates, Spanish surname males 
would have had over 25 percent more deaths frc«n major circulatory diseases. 
By 1980, a fair amount of convergence of death rates from these causes is 
noted. Due to the faster decline in death rates among non-Spanish surname 
males, by 1980, additional deaths from major circulatory diseases would have 
been only about 16 percent of total expected deaths due to these causes. 
Deaths from major circulatory diseases accounted for about two-thirds of all 
deficit deaths in 1970 but only one- third in 1980. At the same time, deficit 
deaths from all cancers rose fron 27 percent of the total to 46 percent. 

The result of these largely offsetting patterns in causes of death is 
shown in Table 3 as the "net difference" between excess and deficit deaths. 
Altogether, the net effect of substituting non-Spanish surname death rates is 
very small because of the nearly equal magnitude of excess and deficit deaths: 
Spanish surname deaths would have been reduced only 2.5 percent in 1970 and 4 
percent in I98O. In those causes in which there is excess or deficit 
mortality the effects are very clear — i.e., either a cause contributes to a 
net excess or a net deficit. Only in the case of cerebrovascular diseases do 
age specific death rates almost exactly offset one another. 

Effects of Mortality Differences on Life Expectancy. - We have 
reviewed mortality levels in the Spanish surname and other white male 
population of Texas and their implications for excess mortality in the former. 
The results are straightforward, and in some cases the excess mortality that 
might be avoided (or the "deficit" mortality that might be added) is quite 
large. Numbers of lives that might be "saved" if death rates were equalized 
have dramatic appeal. But what significance do they have for length of life? 
Life expectancy provides a practical means for illustrating differences in 
mortality of populatio!:is . For most ages (and over all ages when considering 
some causes of death) death rates are so small that, in comparing populations, 
large percentage differences may be statistically significant, but practically 
and substantially trivial in terms of the actual effect of those differences 
on average length of life. In this section we show how differences in age and 
cause specific death rates contribute to differences in life expectancy. 

As an overview, life expectancies at various ages appear in Table 4. 
In 1970, the life expectancy of Anglo males at birth exceeded that of Spanish 
surname males by .4 (68.1 years as canpared to 67.7 years). By 1980, E(0) had 
increased 2.5 years for both groups of males (to 70.5 and 70.2 years, 
respectively) so that the differential remained constant at .4. In general, 
the computations displayed in Table 4 show that, in both 1970 and 1980, 
Spanish surname males have somewhat lower life expectancy than Anglos at 
younger ages, while the reverse is true at older ages. In their analysis of 
1970 data for California, Schoen and Nelson (I98I) report a similar crossover 
to more favorable Spanish surname life expectancy by age 40. The crossover 
occurs even earlier — in fact, by age 25 — in Texas. The maximun life 
expectancy advantage (nearly one year of life) of Spanish surname males is 
reached in middle age, while that of other white males (only .4 years) occurs 
at age zero. The timing of death by cause must account for these patterns. 

Although useful for general descriptive purposes, the canparisons in 
Table 4 are far too broad to permit an adequate assessment of the 
differentials that separate the two ethnic populations of interest. 
Consequently, as notea in the earlier section on methods, we prepared life 
tables for a large number of specific causes of death as a basis for computing 
cause-adjusted estimates of life expectancy. The adjusted life expectancies 



54 



are those for Spanish surname males tnat would result if their age-specific 
death rates for a specific cause of death were to become the same as the rates 
for that cause for -cneir Anglo counterparts. Results of the application of 
this procedure for selected ages appear in Appendix B for the years 1970 and 
1980. The essential findings are shown in Figures 1 and 2, which illustrate 
the difference values in graphic form for selected major causes of death. 
Note tnat since the Spanisn surname life expectancy estimates are taKen as the 
subtrahend, a negative value in Figures 1 and 2 indicates that Spanish surname 
males have an actual mortality advantage, and a positive value denotes a 
current Spanish surname disadvantage . 

In both 1970 (Figure 1J and 19B0 (Figure 2), we see the crossover in 
life expectancy based on all causes of death occurring by age 25 (after which 
the ail-causes slope falls below zero on the Y-axisJ . We also see illustrated 
in botn figures the fact that Spanish surname males are considerably better 
off than tneir Anglo counterparts with respect to the major chronic and 
degenerative diseases. In 1970, Spanish surname males would nave lost about 
one year of life if their death rates from major circulatory diseases had been 
the same as tne Anglo rates (as indicated by values near -1 .0) up until age 
55 > followed by a moderate upward inflection of the curve at older ages. This 
advantage was diminished (to around -.4) and the curve was nearly flat for 
death from circulatory diseases in 1980. 

In regard to cancer also, tne Spanish surname group has more favorable 
death rates, and their relative advantage was greater in 1980 than in 1970. 
The bulK of tnis advantage is due to lower death rates from lung cancer at 
botn points in time. The disparity begins to diminish after about age 45? and 
life expectancy differences related to cancer become extremely small at the 
oldest ages. 

In sharp contrast are the patterns of death due to external causes. 
Spanish surname males nave much higher deatn rates than other white males from 
both motor vehicle accidents and homicide, beginning at birth and lasting into 
the young adult years. After age 30, however, a downward trend in zhe 
differences in life expectancy is clearly evident, indicating a convergence 
with Anglo rates at older ages. In 1970, both homicide and motor vehicle 
accidents contributed very heavily to Spanisn surname mortality, but by 1980, 
homicide was tne major factor depressing life expectancy relative to other 
white males. 

Figures 5 and 4 show 1970 and 1980 intraethnic changes in life 
expectancy adjusted in a manner analogous to the adjustment of interethnic 
differences as found in Figures 1 and 2. (Appendix C presents these results 
in tabular form.; The intraethnic adjusted differences may be interpreted as 
measuring the effects on life expectancy of age-specific mortality gains or 
losses in 1980 as compared to 1970. Positive differences indicate the result 
of lower death rates leading to greater life expectancy and negative 
differences denote higher mortality rates in 1980 than in 1970. The 1970-1980 
differences for Spanish surname males appear as Figure 5, and those for Anglo 
males are found in Figure 4» 

Among Spanish surname males, great improvement is apparent with 
respect to major circulatory diseases (Figure 3j . For all ages up to 75, the 
1980 death rates, had they occurred in 1970, would have produced a gain in 
life expectancy of well over one year. By contrast, there was no gain in life 
expectancy from reduction in the age-specific death rates from external causes 
among Spanish surname males between 1970 and 1980. Improvements with regard 
to deaths from motor vehicle accidents and other external causes were 
competely offset by rising death rates from homicide. 

55 



Figure 1 Differences between observed life expectancy by age for Spanish 

surname white males & tnose wnicn would hd.ve occurred if such raales had 
experienced the age-specific death rates of other v/hite males, for selected 
causes of death — Texas: 19 70 






1.0- 



D 
I 
F 
F 

E 
R 
E 
N 
C 

t: 
I 

N 



I 
F 
E 

E 
X 
P 
E 
C 
T 

u 
c 

Y 



-£#1- 



B---.H. -B, 






O.S-V— ^^-¥"-v., \ 



^w—^~~^ - 



^x \ 



H k, Y 



0.0-i 




\ \ 






■lrh-^*:;:*==*|c=*^4^^ 



\ 



\ 






.^<' 



A-^' 

':^' 



T3- 



^,-*' 



1 .oJV-^-H— K-..^,_^_ 




-1.5- 



>«-^-5<— *<-- ->^' 



■jTT-rrp-rt-rp J i tj 1 1 ttti TTT|-rm-|-rrTT-j tt i i | m rrj-t i i i [ i n i [ i i i i | i i i i ] i i i i [ i i i i | i i i i | i i H ] 

1 122334455667788 
0505O505O50S050505 



AGE 

LEGENDS CRUSE «s.-H^-*- ALL CRUSES 

,*,-*--& LUNG Cfl^jCER 
fi-c>-a EXTERI'^RL 
v~v--¥ HOMICIDE 



#~<^"^ ALL CANCERS 
*<-H->« CIRCULATORY 
-•— I--H MOTOR VEHICLE 



56 



Figure 2 — Differences between life expectancy by age for Spanish surname 
white males & those which would have occurred if such males had experienced 
the age-specific death rates of other white males, for selected causes of 
death — Texas: 1980 



D 
I 
F 
F 

E 
R 

E 
N 
C 

E 

I 

N 

L 
I 

F 
E 

E 
X 
P 
E 
C 
T 
ft 
N 
C 
Y 




*n' "^^^i^.. 






c:Z^-f>- 



-0 . SH 






-1.0- 



1 

-* 

; 



*—■**—-*-—*»---«- 



-^►--i»— « 




-1.5- 



Wt I r I J T I I I J ITT tjTTt ; TTT rrtTT-rTTTTTTTTTT T TT 1 T I T I T I T | M T TTTIII I I rrTyTITI rir H TIIIIliniT" 

1 122334455667788 
050S0505Q50S05Q505 



LEGEND: CAUSE 



AGE 

■H-4-^ RLL CAUSES 
.?,-*-& LUND CANCER 
fi-c^a EXTERNAL 
¥-»i---r' HDMOCICE 



«--«•-# ALL CANCERS 
^♦-H-^ CIRCULATORY 
-H-J-H- MOTOR VEHICLE 



57 



Figure 3 — Differeiices between observed life expectancy by age for Spanish 
surname white males in 1970, and those which would have occurred if such 
males had experiences the age-specific death rates for 1980, for selected 
causes of death — Texas 



D 
I 
F 
F 
E 
R 
E 
N 
C 
E 

I 

N 

L 
I 
F 
E 

E 
X 
P 
E 
C 
T 
fi 
N 
C 
Y 








?=^- 



-0.5- 



■ 1 . G- 



-1.5 




-TTTTTT-rT 



TT7 tirtT TT'I 1 TlTt rTT m 



|iTtrp 



!1-ITTrTTTTT 



1 12 2 3 3 4 4 5 



6 6 7 7 8 8 



5 5 3 



050505050505 



AGE 



LEGEND: CAUSE 



*■ FILL CAUSES 
A-^--* LUNO CANCER 
i3-t>-a EXTERNAL 
v-v-¥ HOMOCIDE 



^-m^^ ALL CANCERS 
*f-H-^ CIRCULATORY 
-1— 1--»- MOTOR VEHICLE 



58 



Figure 4 — Differences between observed life expectancy by age for non- 
Spanish surname white males in 1970, and those which would have occurred 
if such males had experienced the age-specific death rates for 1980, for 
selected causes of death — Texas 



D 
I 
F 
F 
E 
R 
E 
N 
C 
E 

I 

N 

L 
I 
F 

£ 

E 
X 

P 

E 

■r 
I 

N 
C 
Y 




-X— )f— X---H 



._K H — K — 



.* — 



-~^- 



,-.*-- 



1.5h 



n.s 



Q 



"2 
-i 

-i 










-0.5-^ 



-i 
■l.OH 



-1.5- 



j-rrrTT TtrrTTTT i|tttti ttt TiTTt TTT-rit T ri tT-rrrrrT 



■p-TTTj-m 







1 





1 



3 


3 


4 


4 


5 


5 


6 


6 


7 


7 


8 


8 





5 





S 





5 





5 





5 





5 



ftGE 



LEGEND! CRUSE 



^-^,-^ ALL CAUSES 
*•-&-* LUND CANCER 
s-i>-s EXTERhJflL 
v-v-v HOMOCIDE 



#-*-^» RLL CflNCERS 
^<-^->e CIRCULATORY 
-I— 4--+ MOTOR VEHICLE 



59 



Anglo males in 197O (Figure 4) would have had approximately two years 
added to their life expectancy at all ages up to late middle age, if the 1980 
death rates for circulatory diseases had been in effect. Thus, tne 
substantial gains experienced by Spanish surname males due to lower mortality 
from this cause in 19bO were surpassed by the even more remarkable improvement 
among Anglos. 

Non-Spanish surname white males suffered some increase in death fran 
lung cancer, and because of this, while Spanish surname males had some minor 
improvement in mortality due to all malignant neoplasms, Anglos became worse 
off in this regard, beyond this, there were few changes of note in the Anglo 
adjusted differences. 

CONCLUSIONS 

Spanish surname males in Texas have been shown to have only slightly 
higher mortality and slightly lower life expectancy at birth than other white 
males in both 1970 and 19bO. Moreover, a crossover in age-specific death 
rates is observed so that Spanish surname males achieve higher life expectancy 
than Anglos in early adulthood and thereafter. These results are consonant 
with previous work on Mexican American male mortality which was reviewed 
earlier. 

The most salient conclusion fran a policy standpoint is that lower 
Spanish surname life expectancy at birth is due almost entirely to excess 
mortality fran a limited number of causes of death, sane of which may be 
controlled or prevented. Excess mortality among Spanish surname males is 
greatest from external causes of death, accounting for 57 percent of the 
excess in 1980, up fran 48 percent m 1970. In both years, this excess cost 
Mexican American males on the average one year of life expectancy at birth. 
As of 1980, eliminating excess deaths due to motor vehicle accidents would add 
about .2 year of life expectancy at birth for Spanish surname males — about 
half the overall difference fran other white males. Given the convergence of 
age specific death rates from this cause fran I970 to 1980, this seems 
reasonable. Unfortunately, by 1980, most of the disadvantage from external 
causes was due to hanicide, a cause of death that may be unresponsive to 
planned intervention. Excess mortality resulting fran other external causes 
of death, including motor vehicle accidents, declined by over half during the 
1970s, but homicide increased by two thirds. 

Spanish surname males are already considerably better off than other 
white males with respect to diseases that are common at older ages such as 
cancer and heart diseases. They have a disadvantage in excess mortality from 
diabetes mellitus. Better management of diabetes would result in a reduction 
of mortality directly attributed to that disease and also of mortality due to 
the vascular illnesses with which diabetes is often associated. This should 
further increase the advantage in mortality that Spanish surname males have in 
middle age and beyond. 

The fact that Spanish surname males are disadvantaged mainly fran 
deaths due to conditions which may be successfully managed (as diabetes) or to 
events which may be prevented (as motor vehicle accidents) or avoided (as 
homicides) suggests their life expectancy at birth might be fairly easily 
improved. On the other hand, it is certainly possible that if intervention 
programs succeeded in lowering the Spanish surname death rates from external 
causes, thereby allowing a greater number and proportion of the population to 
survive to older ages, deaths fran cancer, circulatory diseases or other 
chronic and degenerative conditions would increase, since these tend to be 



60 



diseases of old age. Finally, it is also true that it is mucn more difficult 
to add significantly to the overall life expectancy of populations whose 
expectation of life at birth is already great (Arriaga, 198 j) . 

REFERENCES 

Arriaga, Eduardo E. 1984. Measuring and explaining the change in life 
expectancies. Demography 21 (June) : 83-96. 

Bradshaw, Benjamin S., and Edwin Fonner, Jr. 1978. The mortality of 
Spanish-surnamed persons in Texas: 1969-1971. Pp. 261-282 in Frank D. Bean and 
W. Parker Frisbie (eds.) The Demography of Racial and Ethnic Groups. (New 
York: Academic Press). 

Bradshaw, Benjamin S. , and Edwin Fonner, Jr. 1980. Survivorship and longevity 
of Spanish-surnamed and otner white persons: Texas, border and nonboraer 
regions, Bexar County, 1969-71, and San Antonio, 1950, I960. Unpublished 
report preparea for the National Institute on Aging. 

Bradshaw, Benjanin S. , and W. Parker Frisbie. 198^. The usefulness of census 
Spanish surname and Spanish origin data with vital statistics data. Paper 
presented at the annual meeting of the Southern Regional Demographic Group. 
Chattanooga . 

Buechley, Robert W. 1976. Generally Useful Ethnic Search System: GUESS. Cancer 
Research ana Treatment Center. (Albuquerque, N.M.: University of New Mexico). 

Davis, Cary, Carl Haub and JoAnne Willette. 1983. U.S. Hispanics: Changing the 
face of America. Population Bulletin 38( June) : 1-43. 

Ellis John M. 1959. Mortality differences for a Spanish-surname population 
group. Southwestern Social Science Quarterly 39(March) :314-321 . 

. 1962. Spanish-surname mortality differences in San Antonio, 

Texas. Journal of Health and Human Behavior 3(Sunmer) : 125-127. 

Gillespie, Francis P., Andrew M. Greeley, Michael Hout, and Teresa A. 
Sullivan. 1983. Public policy, ethnic codes and Hispanic vital statistics. La 
Red/The Net 70( July) :9-13- 

Hernandez, Jose, Leo Estrada, and David Alvirez. 1973. Census data ana the 
problem of conceptually defining the Mexican American population. Social 
Science Quarterly 54(March) :671-87• 
Kautz, Judith A., Benjamin S. Bradshaw, and Edwin Fonner, Jr. 1981. Trends in 
cardiovascular mortality in Spanish-surnamed, other white, and black persons 
in Texas, 1970-1975. Circulation 64 (October ): 730-735. 

Keyfitz, Nathan. 1977. What difference would it make if cancer were 
eradicated? An exanination of the Taeuber paradox. Demography 
14(November):411-l8. 



61 



. 1982. Population Change and Social Policy. (Cambridge, MA: 

Abt Books) . 

Kitagawa, Evelyn and Philip M. Hauser. 1973- Differential Mortality in the 
United States. (Canbridge, MA: Harvara University Press). 

Lee, Eun Sul, Robert E. Roberts, and Darwin R. Labarthe. 1976. Excess and 
deficit lung cancer mortality in three ethnic groups in Texas. Cancer 
38 ( December ) : 255 1 -t)6 . 

Nelson, Verne and Marion Collins. 1976. Computer replication of census Spanish 
surname coding. Mimeo. Berkeley, CA: California department of Health. 

Roberts, Robert E. and Cornelius Askew, Jr. 1972. A consideration of mortality 
in three subcultures. Health Services Reports 87 ( March) : 262-70. 

Rosenwaike, Ira. 1983. Mortality among the Puerto Rican born in New York City. 
Social Science Quarterly 64 ( June) : 375-85. 

Schoen, Robert and Verne £. Nelson. 1981. Mortality by cause among Spanish 
surnamed Californians, 1969-71. Social Science Quarterly 62 ( June ): 259-74. 

ShryocK, Henry S. , Jacob Siegel, and Associates. 1973- The Methods and 
Materials of Demography. Revised edition. (Washington, DC: U.S. Government 
Printing Office) . 

Stern, Michael P. and Sharon P. Gaskill. 1978. Secular trends in ischemic 
heart disease and stroke mortality from 1970 to 1976 in Spanish-surnamed and 
other wnite individuals in Bexar County, Texas. Circulation 
58 ( September ) : 537-43 . 

Sullivan, Teresa A., Francis P. Gillespie, Michael Hout, and Andrew M. 
Greeley. 1983- Surname versus self-identification in the analysis of Hispanic 
data. American Statistical Association, Proceedings of the Social Statistics 
Section, pp. 117-22. 

Sullivan, Teresa A., Francis P. Gillespie, Michael Hout, and Richard G. 
Rogers. 1984a. Alternative estimates of Mexican-American mortality in Texas, 
1980. Social Science Quarterly 65 ( June) : 609-17- 

Sullivan, Teresa A., Francis P. Gillespie, and Richard Rogers. 1984b. Effects 
of ethnic classification on apparent life expectancy: The case of Texas in 
1980. Paper presented at the annual meetings of the Social Statistics Section, 
Joint Statistical Meetings, Toronto. 

Tsai, Shan Pou, Eun Sul Lee, and Robert J. Hardy. 1978. The effect of a 
reduction in leading causes of death: potential gains in life expectancy. 
American Journal of Public Health 68, No. 10 ( October ): 966-97 1 • 

Tsai, Shan Pou, Eun Sul Lee, and Judith A. Kautz. 1982. Changes in life 
expectancy in the United States due to declines in mortality, 1968-1975. 
American Journal of Epidemiology 116, No. 2:376-384. 



62 



U.S. Bureau of the Census. 19 ^2. General social ana economic characteristics. 
1970 Census of Population. PC(1)-C45. Washington, D.C. : U.S. Government 
Printing Oi'fice. 

U.S. National Center for Health Statistics. 1983. Births of Hispanic 
parentage, 1980, by Sephanie J. Ventura. Monthly Vital Statistics Report 32, 
No. 6 Supplement (September). 



63 



APPEWDiX A 
Cause of Deatn Categories ana Corresponaing ICD Codes 
Primary cause ol' deatn 8tn revision 



9th revision 



Maiignant neoplasms, all sites 
Trachea, bronchus, and lung 
All other malignant neoplasms 

Diabetes mellitus 

I'lajor circulatory diseases 

Major ischemic heart diseases 
Acute myocardial iniarction 
Chronic ischemic heart disease 
Cerebrovascular disease 
Pneumonia ana influenza 
External causes of deatn 
Motor vehicle accidents 
homicide ana legal intervention 
All other external causes 



All other causes of death 



140-209 

162 
I4O-I0I , 
1dJ)-209 

2p0 
41 0-41 i, 
4:50-4^8 
41 0-41 :) 

410 
411-41:) 

4:>0-43B 

470-48b 

E800-E999 

E810-E823 

E9bO-E978 

E800-E809, 

E824-E959 , 

E97^-E999 

Residual 



140-209 

152 
I4O-I6I , 
163-209 

250 

410-414, 

430-4:^8 

410-414 

410 

411-414 

430-438 

480-487 

E800-E9y9 

E81 0-E825 

E960-E978 

E8OO-E8O9, 

E820-E939 , 

E979-E999 

Resiaual 



64 





1 Ll 






00 


IV 


T OJ 


O fO 


T T 


il 11 


fj ro 


-H O 


If) 0^ II 




1 a o 




ID 


■•a 


-H 


o 


^ 


'H lf> lO OJ 


-H (VJ 


o o o o 


O O II 




1 -J 




l\. 


t 


■ 


I 




I I 


I ■ 






■ ( 






1 3 VI 






o 


o 


o 


o 


o a Q o 


O O 


o o 


O O 


O O II 




1 O Ul 






1 


1 


1 


1 


1 


1 1 


1 1 


1 




1 1 11 




1 3U1 




























1 3 




























1 xc 




























1 U (J 




























1 1-4 






iH 


*H 


T- 00 


fvl OJ If) T 


r^ ffi 


ro o 


T OJ 


\j) ro 11 




1 xa 




O 


flO 


Ifl lO 


fH 


OJ T 


ro OJ 


'^ ° 


oo 


o a 


O 'H II 




1 ZLUi 




yO 


■ 


■ 


■ 


f 


I • 


• ) 












1 1- 






o 


ooooooooooaoaoo ii 




1 Ul u 






1 


1 


1 


1 


1 


1 1 


1 1 






1 1 II 




! l/> UJ 




























1 O -J 


1 tH 


























X UJ 


1 OD 


























(- in 


1 1 




























1 <f- 




If) 


ro 


5V 


T 


T- T- 


If) O If) o 


T ro 


00 OJ IV O II 




o oc 


\ IS. 


Ifi 


rv. 


IV T <M (M T T Kl 


t ° 


O -H 


O-H 


O O 11 




z o 


1 o\ 


T 


■ 


■ 


I 


I 


1 f 


■ ■ 












(Z u. 


t «^ 




o 


O 


o o o o 


o o o o 


o o 


d d 


d d 11 
1 II 




1/1 n 






1 


1 


1 ' ' 






I 1 ' 










Ul u\ 




























-J Ul 




























(E -1 1 




























E C 1 






<M 


IT 


O M5 


(<■) If) f\. <-< 


vf) rH 


T OV 


ro fO IV OJ II 




E 




O 


fO 


(V 


Ifl OJ 


OJ T 


T ro 


-H O 


O T 


-H ^ 


O »^ II 




Li 




tr> 


■ 






• 
















1- U 






O 


O 


o 


o 


o a 


d d 


d d 


d d 


d d 


d d II 




M 1- 






1 


1 


1 


1 


1 


1 1 


1 






1 II 




X M 1 




























a X 1 




























3 1 




























Ul 1 




























z ae 1 






ro 


1-1 


IV Ul 


r-j -H 


T OV 


If) -H 


If) T 


CD O 


If) <jv II 




CUJ 1 




a 


T 


IV. 


T 


OJ 


OJ T 


^ OJ 


•-I a 


o o 


n a\ 


O fO II 




Z X 




























QC 1- 1 






O 


o o o 


D d 


d o 


d o 


d -H 


d d 


d d II 




3 O 1 








1 


1 


1 


1 


1 1 


1 






1 II 




Ln 1 




























U. 1 




























X a 1 




























w 1 


i H 


■i H 


H 






















M Ul 1 






10 


OJ 


ro 


»H 


ii5 a> 


03 ffv 


rt If) 


OJ -< 


•H o 


N OV II 




Z U 1 




m 


T 


o 


o o 


O \D 


T OJ 


OJ OJ 


O o o o 


° 1 " 




ec k- 1 




IS. 


1 


■ 


I 


t 


■ 1 


t • 


t 1 


t ■ 


■ ■ 






a. o: 1 






o 


aOOOOOODOOOOOOQ II 




in oc 1 






1 






1 


1 


1 1 


1 1 


1 




1 II 
ii 




U X 1 


























O ►- 1 
























II 


OD 


li. <S 1 
























II 




Ul 1 






»■* 


» 


*-< 


N. 


IV fy. 


T -- 


T If) T (\J T OJ 


T T 11 


X 


m O 1 




o 


M3 


OJ (M O 


'H IV 


rv If) 


OJ O 


o o 


o o 


O OJ II 


w 


Ul 1 




t£ 






■ 


1 












> • II 


a 


(9 U 1 






Q 


ooooaoooooooo 


O O II 


z 


X l-l 1 






1 


1 


1 


r 


1 


1 1 


1 1 






1 II 


iij 


U. 1 
























II 


a. 


a M 1 


iH 






















II 


0. 


UJ U 1 


1 l\. 






















II 


c 


l-H Ui 1 


1 1 






















II 




h. a. 1 


o\ 




Ifi 


T- 


oo 


rt 


ro ^ 


If) OJ 


If) OJ 


T OJ 


rH ro 


rt gs II 




M l/l 1 


•n 


\o 


^ 


T 


ro 


^ 


00 o 


o r>^ 


lO O 


o o 


rt o 


-H OJ II 




U 1 1 


a\ 


T 


t 


1 


1 


t 


1 I 




■ t 






• 1 II 




Ul Ul 1 


rH 




O 


o 


o o 


O -H 


-H O 


o o 


o o 


o d 


O O II 




0. <S 1 






1 


1 


1 


1 


1 


1 1 


1 






1 II 




Ifl X 1 
























II 
1 
1 




1- UJ 1 


























X X 1 
























1 




1- 1 






m 


OJ 


(TV 


ro 


T \0 


i-" \i3 IV ro 


T (V. 


05 -H 


OJ fO II 




% 1 




O 


\D 


IT 


ro 


T-l 


OJ o 


-^ IV 


fO o 


O OJ 


1 OJ 


-H XT 11 




U O 1 




(O 






• 


1 


■ 1 




■ I 




■ ■ 


■ • 11 




Z UJ 1 






a 


o 


o 


o 


o -< 


■H a 


o o 


d d 


o o 


O O II 




X U 1 






1 


1 


1 


1 


1 


1 1 


1 






1 II 




1- Z 1 
























II 




U Ul 1 
























II 




Ul l-H 1 
























II 




0. oc 1 
























II 




X Ul 1 






i-t 


(J\ 


\0 


ro 


lO IV 


ro o 


T ro 


T -H 


\D a 


If) \£ II 




UJ 0. 1 




o 


T 


T 


m 


1-1 


OJ jv 


o rv 


ri") a 


-H J\ 


ro If) 


O \fl II 




3C 1 










I 


« 






■ ■ 






■ ■ 11 




Id UJ 1 






a 


o o 


o 


D O 


rt o 


o o 


o o 


d d 


O O II 




h. 1 








1 


1 


1 


1 


1 1 


1 






II 




M a 1 
























II 




-1 X 1 
























11 




X 1 
























II 




O 1 


H 


•i H 


•1 


















II 




U L/l 1 
















I ■ 








in II 




=> UJ 1 




^^ 










lyi 


m u 








UJ II 




oc _l O 1 




X 










UJ 


H u 


Ul 




i/i 


in II 




Ul X <S 1 




^^ 










1^ 


a X 


IJ\ 




1- 


3 II 




un r cr 1 




u 




m 






X 


u. 


X 




z 


X II 




00 rt 1 




15 




z 






Ul 1- z 


Ul 




Ul 


(J II 




O X 1 




X 




m 






m 


g" 


in 


X 


a 


II 




U « 1 








X 






t-l 


n 


l\l 


HH 


_i II 




Z = 1 








_l 






a 


U -J 


a 


z m 


U 


X II 




UIW O 1 








0. 








X X 




UJ UJ 


u 


Z II 




u r<. 1 








o 






u^ > 


H-l 


• c^ 


3 in 


X 


Oi II 




3 U. OV 1 








UJ 






3 oc 


u a 


O X 


_l 3 




u in II 




1- M •-< 1 








z 






1- o 


11 oc 


■ _l 


Lu X 


u 


1- Ul II 




Ul 1 




X 










M ^ 


E X 


I 3 


z o 


-J 


X n II 




m a •• 1 




1- 


in 


1- 






-1 X Ul u 


■ u 


n 


Ij 


U 3 II 




UJ in 1 




X 


UJ 


z 






_l _l 


X o 


11 Ln 


_l 


n 


X II 




in QC X 1 




Ul 


in 


X 






UJ 3 


O 3- 


X 


<a X 


X 


a: u M 




Ul Of X 1 




a 


3 


z 






z u 


in E o 3. 


z 


Ul Ul 


U II 




U =) UJ 1 






X 


o 






oc 


n 


11 o 


X QC 


^ Q 


X oc II 




Z U )- 1 




u. 


u 


f-( 






m 1-1 


UJ 


z oc 


H UJ 


n 


1- Ui II 




UJ U 1 1 




o 




_l 




a: 


UJ tJ 


QC 1- 


O OJ 


Z 1- 


o: u 


O X II 




o: o 1 1 






-1 


X O Ul 


1- 


D 3 


q: u 


O X 


O n 


V- II 




Ul Z 1 




Ul 


_l 


E 


z 


X 


uj oc 


-) U 


X ct: 


C UJ 


1- E 


_l O II 




U. Ul 1- 1 




in 


X 




3 


1- 


CO o 


X X 


o u 


3 


o o 


_i II 




U. ^ X 1 




3 




_1 


_l 


o 


X -) 


E 


u 


U _1 


E X 


X -1 II 




M S UJ 1 




X 




-J 






li X 






Z -1 




-1 II 




a X o 1 




o 




X 






O E 






0. X 




X II 



65 



Q 
Z 
Ul 

a. 
a. 

(Z 



a 
z 

Q UJ 



in u 
u o 



ji 
II 
II 
II 
II 
II 
II 
11 

M 

II 
II 
II 
II 

II Ui 

II I- 
II i-i 

II X 

II 3 
II 

II OC 



II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 

II UJ 



a 

UJ » 
)- IT- 

M »-' 

X 
3 K 

o 
a: u. 

UJ 

X 01 
I- Ul 
o k- 
c 
a a 
z 

X X 

^- 

UJ E 

r UJ 

X a 
z 

oc u 

3 M 

vt u. 

X u 
in UJ 

h-l 0. 

z m 

X I 
Q. UJ 

X 

tt: 

O Ul 
Li. X 

I- 

in 
u a 

15 UI 

s u 

z 
a UJ 

U M 

M oc 

U. Ul 

hH a. 
u X 

UJ UJ 

a. 
Ul a 

X 
I- X 
X 

tn 

> UJ 
o _J 
z X 
X X 

I- 

L) X 

UJ u 

0. 3 

X l/l 
Ul 

U. 

Ul hi 
L. 

i-i a 

-I UJ 

Of 

Q Of 

Ul 3 
i U 

q: u 

Ul o 

in 

ffi UJ m II 

Z X U II 

Ul »- II 

Ul O I II 

3 -I I II 
1-3X11 

UJ O I- II 

CO 3 X II 

UJ II 

m X a II 

Ul (J II 

O HH U. II 

Z X O II 
Ul 3 II 

Qc in II 

UJ Ul UJ II 

u. Ul in II 

U. O 3 II 
h-i X X II 
Q t- (J II 



II : 

II i 

II c 

II 

II 

II 

II 

II 

II 

II 

II 

II 

II 

II 

II 

M 

II 
II 
II 
II 

M 
II 
II 
II t-H Hi 



II OC 

II 3 

II in 
II 

II X 

II m 

II M 

II z 
II X 
II 0. 

II in 

II 

II 

II 

II 

II 

II 

M 

II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 
II 



<\i o i\. po (NJ T fo oj CO 10 \fi ro --I O (\) \0 

O OOOO'HOOOOOOOOOO 

III I 



IV -'(Nj'^ror^ivjo^a^n.ooflOTOTCJ 

r^ rti-IOOsflrjvOTrOOOOOOT 

t ■■tllltlltlllll 

■H oooo'-i'-iaoooooooo 



lf» 
l\. 



If) 



iN-of)rorj««Ti\.o\or^'<oso 



»-t 


o o o 
1 1 


O <M 


ridao 


do 6d 6a 
1 1 




^oflo^olot^.(\JfOT^^.olnlo<^J(^Jffl 

00000\OQ10n'H'^00'-l(M 


fH 


o o o 
1 1 


O (\J 


rt rH O O 


o o o o o o 
1 1 


in 

in 


T-i a, CD 
a a a 


O T' 


T rv. '^ vfl 
in <T* in M 


o so "-I r*. M r. 
(vj Q o o -« ro 


OJ 


o o o o 
1 


^ 


-^ooooooooo 


in 


T (\l \0 
O O O 




rtco'^fjcousrjom'H 

>£PJ(0P0OOOOOP0 


o 


O O O 
1 


o a o o o o 

1 


o o o o o o 
1 




lnTO^(^Jff'M(SI»fSI 

ooooMOorroT 


-< "^ in o in r^ 

rt rH o o o w 


T-l 


o o o 

1 


o 


*H 


o o o a 


o o o o o a 

1 


in 


ao in po 
o o -< 


o 


T 


in \fl ru (M 
o\ m lo T 


rt IV ^ o\ in r. 

r^ Q tH O O "^ 


«-( 


o o o 
1 


o 


■H 


o o o o 


o a o o o o 

1 1 


o 

\0 


CM T \0 

W O 'H 


o 




T (V. sC <M 

o\ in to T 


rt (f) W ^ (\. (\. 
rH O 1-1 <\J O O 



0000<-l0000000000 

I III 



poMsflcoMoororooD'-f-HfMoiio 
•-•O'-'OMooinM^roarjTfjin 

OOOO'^OOOOOOOOOO 

I I I 



I = I 



X 
Ul 

a 



u 
m 

3 
X 
O 



Ul 

in 

3 
X 
U 



-I 

X 



Ul 

E 

i£ 
-I 
Q. 
O 
UJ 

z 

1- 
z 

X 

z 



hi oc 

□ X 
u. 

i- z 



If!*' 



m > 
3 oc 
I- o 
M t- 
-I X 
-I _i 

UJ 3 

in M 
oc UJ u 
(S UJ I- 

z X UJ oc 

3 I- CO O 

_1 O X -5 

l-l X 

C3 X 



X X 

h-l • OC 

u a o X 

M oc • _i 

E X X 3 

U U • U 

X o 1-1 in 

U 3- X 

Ul E (J =• 

1-4 1-4 O 

UJ Z QC 

U I- O CD 

D 3 U Ul 

T (J X OC 

X X U UJ 

E U 



15 

Z Ul 
Ul Ul 
3 Ul 
-I 3 
U. X 
Z U 
M 

_l 
o» X 

z 
X a: 

1-1 Ul 

z ^ 

O X 

E U 
3 

U -I 
Z -J 
Q. X 



Ul 

o 



in 

UJ 

in 

3 
X 
U 



UJ 



X 

z 
a: 

Ul Ul 
I- Ul 
X Ul 

U UJ 3 
1-1 X 

X oc u 

UJ U Ul 

2- a X oc 

h-i I- UJ 

a: (J o X 

O M ^ 

I- E -J o 

O O _J 

E X X -J 



66 



The Contribution of 
Socioeconomic Position to 
Minority Health 




Mary N. Haan 

George A. Kaplan 

Human Population Laboratory 
California Department of Health Services 
Berkeley, California 



The Contribution of Socioeconomic Position to Minority Health 



Introduction 

Investigation of the differential health experience of minorities and 
whites cannot help but raise important questions concerning the reasons for 
these differences. In Part 1 of this report, we will argue that socio- 
economic position (SEP) represents an important and plausible area of 
investigation in the search for reasons. It is important because SEP and 
minority status are clearly intertwined, and examination of both will 
potentially clarify our understanding of minority health. We say plausible 
because limitations in the available data advise caution in interpretation 
and application. However, there is considerable information which points 
to the critical role of SEP. We will examine this role in several stages. 
First we will review the strength and consistency of the association 
between SEP and a variety of disease outcomes. Rather than focusing on 
specific organ systems, we will use the epidemiologic triad of person, 
place, and time to organize the massive amount of evidence on SEP and 
health. Then we will consider the association between SEP and membership 
in minority groups. Our next step will be to consider available evidence 
concerning the consistency of the association between SEP and health, 
between and within specific minority groups. We will then move to evidence 
which indicates how much of the differences in health between minorities 
and whites can be attributed to SEP. In Part 2, we will examine evidence 
which suggests some of the ways in which low SEP may be associated with 
poorer health. 

As a first step, we need to consider for a moment what is meant when 
we refer to SEP. This topic has been addressed recently by Morgenstern 
(66). Most investigators have viewed SEP as an amalgam of income, educa- 
tion, and occupation. Various indices have been constructed in an attempt 
to combine, on empirical or theoretical grounds, information from these 
three domains. Indices of social status have also been constructed, again 
on empirical or theoretical grounds, which rank people according to 
"prestige." Lastly, the construct "social class" has been used to order 
groups in a number of ways, ranging from broad occupational groupings to 
orderings based on influence, authority, and power in the economic 
structure. 

It is clear from this brief discussion that we can mean many things by 
SEP. Measures which combine different domains of socioeconomic information 
can hinder our understanding of the ways in which SEP is associated with 
health. Although different socioeconomic measures may be related, they 
have differential utility depending upon the question being asked. For 
example, in situations where illness is likely to effect occupation and 
income, education may be the preferred socioeconomic measure. Income 
level, however, may be more important in obtaining services or meeting 
needs than education would be. On the other hand, because of secular and 
group-specific trends in educational attainment, level of education may 
behave differently as a risk factor for different cohorts. Measures of 
occupation which group individuals in broad classes such as professional- 
technical, managers-administrators-proprietors, or semiskilled operatives 
may obscure large income or educational differences within these classes. 
On the other hand, occupation may summarize the cumulative effects of 
education and income or measure other aspects of SEP not tapped by 
education and income. 

69 



In what follows, we will report on socioeconomic measures which are 
derived from income, education, or occupation. Where possible, we will 
utilize more than one measure in our discussion. In addition, in some 
cases, we will utilize measures which reflect characteristics of the areas 
in which individuals live. Census tract characteristics such as median 
family income, median years of education, or per cent in a particular 
occupational category are sometimes used as proxy measures of individual 
levels. Recognizing the potential "ecological fallacy" (65) involved in 
the use of such measures, vje will use them with great caution and only 
where the pattern of findings is consistent with results using other 
measures. In Part 2 of this report, we will indicate, however, how such 
area characteristics may, in themselves, be important to our understanding 
of the health differentials between minorities and whites. 

Part 1 

All-Cause Mortality: The General Picture 

Ever since the 12th century, when data were first recorded on this 
topic, those at the lowest socioeconomic levels in the community have been 
found to have higher death rates (1, 90). This pattern is reflected in a 
large number of reports which have examined the association between socio- 
economic factors and all-cause mortality. An illustrative example comes 
from Kitagawa and Hauser's study of adult mortality in the United States in 
1960 (51). They found a consistent inverse gradient of mortality rates 
associated with socioeconomic position. Those who had higher SEP had lower 
mortality rates. This was true whether the measure of socioeconomic 
position was based on family income, median income of census tract of 
residence, education, or occupation. In the many studies of which we are 
aware, this pattern of increased all-cause mortality associated with lower 
socioeconomic position is found in well over 80 per cent. Furthermore, in 
many cases, there is an orderly gradient of rates associated with increases 
or decreases in SEP. In what follows, we will briefly examine the consis- 
tency of this finding for different age groups, diseases, geographical 
locations, and time periods. 

Consistency by Age 

All-Cause Mortality. Socioeconomic gradients of all-cause mortality 
are found in most age ranges. There is some evidence that the association 
between socioeconomic factors and health is somewhat weaker at the older 
ages. Kitagawa and Hauser (51) found that the gradients associated with 
income and education were larger for persons 25-64 years of age than they 
were for those 65 years or older. In analyses (49) of the 18-year mor- 
tality experience of a large (n=6,928) cohort of individuals representative 
of Alameda County, California, in 1965 (6), we found that the increased 
risk associated with low compared to high family income decreases with age, 
becoming non-significant between 60 and 70 years. Others have reported 
similar findings (51). However, in interpreting the significance of this 
apparent dilution of effect, we must take into account the fact that income 
generally declines with retirement, resulting in a disproportionate lower- 
ing of the income of those who were not previously in the lower income 
categories. Thus some portion of those in the lower SEP groups have only 
recently entered these groups. The absence of lifelong measures of SEP may 



70 



result In dilution of the association between SEP and mortality in the 
later years. Similarly, average levels of education have increased in 
successive birth cohorts, and the educational requirements for most occupa- 
tions have increased. It is, therefore, reasonable to believe that the 
health consequences of a low education may have similarly increased. As 
always, the interpretation of "age" effects are complicated by period and 
cohort effects. 

Diseases of the Young. The overall consistency of the association 
between socioeconomic position and health status can be further seen by 
examining outcomes which are age-related. A substantial body of evidence 
exists linking higher rates of infant mortality to socioeconomic position 
(2, 7, 12, 16, 35, 36, 60, 73, 79, 84, 102). Many studies have shown that 
perinatal and infant mortality rates are elevated for those with lower 
income, lower educational attainment, poorer occupational status, or other 
types of social disadvantage (9, 10, 51, 105). There is also evidence 
which suggests that higher rates of birth defects are found in the poor 
(14, 28). The major sources of mortality from unintentional injury in 
children (housefires, drowning, and suffocation) also show a strong asso- 
ciation with SEP (4). A similar pattern is found for a wide variety of 
health outcomes in the young (9, 10, 27, 57, 63, 83, 85, 89, 92, 104). 

Diseases of the Middle Years. When we turn to diseases of middle-age, 
we see similar patterns. Vital statistics data confirm the inverse asso- 
ciation between SEP and the various manifestations of atherosclerotic 
disease (29, 37, 46, 69, 76, 80). In the United States, both prevalence 
and incidence of cardiovascular disease are inversely related to SEP (15, 
48, 50, 51, 53, 78, 95, 112), although for some groups, there have been 
changes in the direction of the association between SEP and cardiovascular 
disease (64). In the 1972 Health Interview Survey (101), those who had 
family incomes under $5,000 had 33 per cent higher prevalence of heart 
conditions than those with family incomes of $15,000 or more. The rates of 
hypertension without heart involvement were over 60 per cent higher in the 
poorer group. Similar findings have been reported in a number of studies. 
Findings from the Hypertension Detection and Follow-Up Program show a 
strong inverse gradient of prevalence of hypertension associated with 
years of education (41). Mortality from coronary heart disease shows a 
consistent SEP gradient when SEP is measured by occupational groupings as 
well (51). In addition, survival from coronary heart disease appears to be 
inversely associated with SEP (82). An SEP gradient is also found for 
unintentional injuries in this age group. For example, residents of low 
income compared to high income counties have about three times the 
mortality rate for motor vehicle occupants even through they are likely to 
be driving less (4). Similar patterns are found for other diseases of this 
age group (11, 31, 40, 91, 112). 

Diseases of the Later Years. Many cancers which reach their peak 
prevalence in the later years show inverse gradients with SEP. Among these 
are cancers of the lung and pleura (56, 81, 109), oral cavity and pharynx 
(42, 111), esophagus (59, 110), and stomach (43, 93). There are, of 
course, sites which evidence the opposite gradient such as breast (3, 26) 
and testicular (68, 77). However, it is notable that the poorer survival 
associated with lower SEP is found both for sites where there is an inverse 
association with SEP, for example, prostate (21), and for sites where there 
is a direct association with SEP, for example, breast (22). Gradients of 
disease related to SEP are also found for stroke, osteoarthritis, and other 
diseases and for various measures of disability and impairment (19, 51, 99, 



71 



100). Mortality rates for those in this group from pedestrian injuries, 
falls, fires and burns, and exposure to cold also show a strong association 
with SEP (4). 

Although not an exhaustive listing, the evidence presented above is 
quite compelling regarding the consistency of the association between 
socioeconomic position and health at different age groups. In general, 
those at lower levels of SEP have higher rates of most diseases, covering a 
wide range of ages and organ systems. 

Consistency by Place 

The association between socioeconomic position and health is consis- 
tently found throughout the world. Mortality differentials associated 
with socioeconomic position are found in countries as diverse as the 
the United States and India. These differentials are found in England and 
Wales (67, 76), Sweden (30), Finland (69, 80), France (24), Norway (45), 
Australia (29), New Zealand (74), Latin America (5), Ghana (13), Canada 
(62), and many others. Within the United States, associations between 
socioeconomic position and health outcomes have been found in such diverse 
places as Evans County, Georgia (95), and Alameda County, California (38); 
Iowa (31) and Hawaii (83); and Chicago (51) and Charleston (50). 

Consistency over Time 

Despite the large improvements in health seen during the last 60 
years, the gradient of health associated with SEP has changed very little. 
Hollingsworth (45) has done the most extensive review of changes in the SEP 
health gradient over time. He examined changes in all cause mortality by 
occupational class in England and Wales for the period 1891-1971. The 
standardized mortality ratio for the lowest social class (V) compared to 
the highest social class (I) in the period 1890-1902 was 1.50. In the 
period 1970-1972, the same ratio was 1.58. Although there clearly are 
problems in the comparability of data sources and definition of social 
class over this 80-year period, the similarity between the two figures is 
striking. It is especially striking when we consider that these data cover 
a period in which there were major changes in the leading causes of death. 
Similarly, Kitagawa and Hauser (51) found very little convergence of the 
socioeconomic differentials in mortality for all causes, excluding infant 
mortality, in Chicago during the period 1930-1960. Others have reported 
similar findings (54, 88). Analyses of mortality between 1960-1970 in 
Birmingham, Buffalo, and Indianapolis suggest that there was a slight 
increase in the gradient of mortality associated with socioeconomic posi- 
tion during that period (112). Some reports have suggested that the 
association between socioeconomic position and specific diseases has 
changed over time (58, 64). For example, mortality from cancer of various 
sites has changed over time (56). Blaxter (8) summarized these changes in 
England and Wales between 1930-1963 by noting that for sites which have 
been more common in those of lower socioeconomic position, the gradient 
associated with SEP has increased, whereas the gradient associated with SEP 
had decreased for sites more common in those of higher SEP. For sites 
which are decreasing in mortality, SEP gradients are increasing, and for 
sites which are increasing, SEP gradients are reversing. 

Changes in the SEP gradient for coronary heart disease have also been 
noted. For example, in analyses in Evans County, Georgia (64), and England 



72 



and Wales (58), there appears to have been a reversal In the SEP gradient 
for mortality from coronary heart disease (CHD). That is, CHD among low 
SEP men has increased, and CHD among high SEP men has decreased. However, 
in both cases, this trend has been seen only for men; low SEP women have 
consistently had higher rates than high SEP women. It is important to note 
that although SEP gradients for CHD in men may have reversed in rural 
Georgia (Evans County) during this period, there is no evidence in the 
total mortality experience of this cohort which suggests that at any time 
during the early part of the 20-year follow-up, low SEP individuals had 
better survival than high SEP individuals (95). 

To summarize, lower SEP is consistently associated with poorer health. 
This association is found when considering different ages and diseases, 
different geographical locales, and has been relatively stable over a 
considerable period of time. In the next section, we will present evidence 
which argues for the important role of SEP as a risk factor in the examina- 
tion of minority and white health differences. 

Socioeconomic Position and Minority Status 

This section describes the socioeconomic position of minority groups 
in the United States. Data on income, education and occupation is 
presented for blacks, hispanics, Asians, and American Indians. Much of the 
available data permits only analysis of white compared to "non-white." The 
"non-white" group is approximately 85 per cent black and 15 per cent other 
"non-whites." When possible, more detailed groupings will be presented. 

From a health standpoint, the lack of detailed information on other 
minority groups is a deficit since what data does exist suggests there are 
some important differences in SEP and in health status between the various 
minority groups. 

Income 



Table 1 shows the income distribution for hispanics, blacks, and all 
others including whites (23, 97). This data shows that hispanics and 
blacks are similar in income and that both have substantially lower incomes 
than whites. For children under 18, black children are four times more 
likely to live in poverty than whites. When the family is headed by a 
woman, black children are 56 per cent more likely to live in poverty than 
whites. However, the poverty status of blacks is not entirely due to the 
higher proportion of female-headed households. In fact, the black-white 
poverty difference decreases when comparing female-headed households only, 
suggesting that presence of dependents and lower incomes afforded women in 
general also serve to increase the poverty rates. 

Comparison of 1970 median incomes earned by non-black minority groups 
shows that white males earned more than three times that earned by American 
Indians, 47 per cent more than Japanese males, and twice that earned by 
Chinese males and Filipino males. The median income differentials were 
less striking for females, but white females tended to earn 10 per cent 
more than other females except for American Indian women, who earned two 
times less than white women (23). 

Occupation 

The white labor force participation rate is 7 to 8 per cent higher 



73 



than the rate for blacks and other minority groups. This picture is 
further complicated by the fact that black women have a 4 per cent higher 
participation rate than white women, while black men have an 8 per cent 
lower rate than white men. Examination of employment status among persons 
over 16 years of age for other minority groups shows that American Indians 
are the most disadvantaged (36% of males not in the labor force), while 
Japanese, Chinese, and Filipino males are similar to whites, with approxi- 
mately 21 per cent of males over 16 not in the labor force. Rates for 
women of non-black minority groups follow a similar pattern, except that 
their non-participation rates tend to be around 50 per cent. About 65 per 
cent of American Indian women are not in the labor force (23, 97). 

Table 2 shows the occupational distribution by minority group and sex. 
These figures were calculated as the relative proportion of whites employed 
in a category to minorities employed in that category. For example, white 
females were 39 per cent more likely to be employed in white collar jobs 
than black females. Consistently, minority groups have proportionately 
fewer members in white collar jobs and greater numbers in blue collar and 
service jobs. Blacks, hispanics, and American Indians are most similar in 
this regard. Asians are more similar to whites, except for employment in 
service jobs where white females exceed Asian females, and white males are 
slightly fewer than Asian males. 

Job tenure also varies by minority status. Thirty per cent of white 
males have job tenure of 20 years or greater. Females of both groups have 
the shortest tenure (10%), and black males have job tenure 10 per cent less 
than that of white males. Job tenure is associated with increased social 
stability, increased income, and increased post-retirement benefits and 
therefore affects socioeconomic position. 

Education 

Blacks, hispanics, and American Indians have lower educational attain- 
ment, and lower college enrollment than whites. This is much less true for 
Asians, whose educational attainment is similar to whites. According to 
1978 data, among persons over 18 years of age, 83.9 per cent of whites, 
69.8 per cent of blacks, and 55.5 per cent of hispanics were high school 
graduates. Blacks aged 18-19 had lower college enrollment (25%) compared 
to whites of the same ages (35%). Hispanics were eight times less likely 
to have college or greater education than whites and 2.5 times more likely 
to have a less than eighth-grade education than whites. 

In 1978, the black-white ratio for college enrollment among persons 
14-34 years of age was .13 for men and .15 for women. Tables 3 and 4 show 
educational attainment levels for whites and minorities. 

The data presented above abundantly demonstrate that blacks, 
hispanics, and American Indians are of lower socioeconomic position than 
whites as masured by income, education, and occupation. Asians appear to 
be at less of a disadvantage with respect to educational attainment but are 
also disadvantaged with respect to income and occupation. The recent 
changes in immigration patterns may have altered the socioeconomic position 
of Asians, and an examination of more recent data (i.e., 1980 Census) could 
be useful. It seems apparent that our understanding of minority health 
status must include examination of SEP. The evidence provided in previous 
sections on SEP as an independent risk factor and on the close association 
between minority status and low SEP point to the need for this approach. 



74 



Minority Status, Socioeconomic Position, and Health 

This section will review research on minority status and health which 
also examines socioeconomic position. We will specifically examine all- 
cause mortality, cardiovascular disease, cancer, infant mortality, and 
mortality from non-disease causes such as accidents, fires, and drownings. 
In general, research on minority and health has not simultaneously examined 
SEP. This is of particular concern because of the close association 
between minority and socioeconomic position discussed above. Without such 
an approach in studies of minority health, especially of the more disadvan- 
taged groups such as blacks, hispanics, or American Indians, it is 
difficult to conclude whether any results obtained are due to some minority 
characteristic or due to the socioeconomic conditions prevailing in that 
group. 

All- Cause Mortality and General Morbidity 

Black, hispanic, and American Indian minority groups in the 
United States generally incur higher mortality rates from all causes and 
exhibit higher rates of other indicators of morbidity. Table 5 shows some 
measures of morbidity for blacks, hispanics, and all others (including 
whites) of all ages for incomes less than $5,000 and greater than $15,000 
for 1976 (106). The prevalence of morbid conditions, hospitalizations, and 
activity limitations are negatively associated with income for all groups. 
The rates for each of these measures of morbidity are very similar between 
the different groups at each income level. The number of days of 
restricted activity varies somewhat but a consistent income gradient is 
still present within each group. 

Kitagawa and Hauser (51) have shown that SEP is consistently asso- 
ciated with mortality and that the association between SEP and all-cause 
mortality is as consistent within minority groups as it is for whites. A 
problem found in the Kitagawa and Hauser study and many others is that the 
great majority of minorities have lower incomes than whites, making 
adjustment for SEP difficult within minority groups and for purposes of 
comparison to whites. 

An opportunity to examine the contribution of SEP to the differential 
survival experience of whites and blacks presented itself in studies of 
Alameda County, California, residents. In 1965, the Human Population 
Laboratory of the California Department of Health Services selected a 
representative sample of almost 7,000 adults to participate in a longitu- 
dinal study (6). The mortality experience of this cohort has been 
ascertained through 1982. Survival of blacks, as expected, was poorer than 
whites. A proportional hazards model showed that the age-sex adjusted 
hazard rate was 34 per cent higher for blacks (p=.004). When a measure of 
income adjusted for family size was introduced, the difference between 
black and white survival was no longer significant, while the impact of 
income was significant (p=.0001). Figures la and lb present the differen- 
tial survival experience of blacks and whites in this cohort without 
adjustment for income (la) and with such adjustment (lb). Thus in these 
analyses, differences in SEP appear to account, to a great extent, for the 
differential survival experience of blacks and whites. 

The association between SEP, minority status, and health is relatively 
consistent when specific disease outcomes are examined. The next sections 
will discuss these associations with respect to cancer, cardiovascular 



75 



disease, and infant mortality. These outcomes were chosen because they 
represent major causes of morbidity and mortality and because data are 
available that permit adjustment for both minority status and SEP. 

Cancer 

Several studies have reported associations between increased cancer 
incidence and poorer survival with socioeconomic position (17, 55, 71, 72). 
The association has been observed for cancer incidence and survival for 
all sites but varies by specific site. Similarly, differences in cancer 
incidence and mortality vary by minority group. For example, blacks have 
higher incidence rates and poorer survival from rectal cancers than whites. 
White women have higher incidence of breast cancer and better survival than 
black women (71). 

A study of cancer patient survival among minority groups in the United 
States reported that survival from all-site cancer was substantially worse 
for blacks, American Indians, and Chinese than for whites (113). Table 6a 
shows the ratio of white five-year survival rates to each minority group's 
survival rates for males. Table 6b shows these ratios for breast and corpus 
uteri for females. 

Black males and Hawaiian males have been reported as having higher 
cancer incidence rates than whites and other minority groups for all sites. 
(107). White females and black females have similar incidence rates (300/ 
100,000) for all cancer sites, while Hawaiian women have much higher rates 
(400/100,000) than all other groups. Site-specific incidence rates for 
blacks vary considerably, as do those for other minority groups. For 
example, black males have higher incidence rates than white males for lung 
cancer, pancreatic cancer, and prostatic cancer, and black females have 
higher incidence rates than white females for cervical cancer. Hispanics 
have notably lower incidence rates than most other groups for all sites and 
for most site-specific cancer incidence rates. Although site-specific 
cancer incidence among blacks is not dramatically higher for most sites 
than white rates, survival from cancer for blacks and some other groups is 
poorer than whites' for many sites. 

Blacks' five-year cancer survival is poorer than whites' for colon, 
rectum, nasopharynx, larynx, lung, bronchus, skin melanoma, prostate, 
urinary bladder, kidney, pelvis, brain and other nervous system, thyroid, 
non-Hodgkin' s lymphoma, breast, and corpus and cervix uteri. American 
Indians also have poorer survival rates than whites for a large number of 
cancer sites. In general, blacks and American Indians are at greater risk 
than whites, while other minority groups appear to do better than whites. 
Notably, Chinese do worse with all-site and stomach cancer than whites. It 
is useful to note that the more disadvantaged groups (i.e., American 
Indian, black) have poorer survival rates than whites for many sites (113). 
SEP may also affect survival from cancer by affecting access to medical 
care or availability of information on cancer. Unfortunately, data showing 
incidence and survival for each minority group by SEP are not available. 

Cancer incidence from some sites may also be associated with SEP. For 
example, a study of coke plant workers found that blacks had a lung cancer 
SMR six times greater than whites employed in the same plant (61). Black 
workers in this study were employed in much greater numbers in jobs where 
the exposure to benzopyrene and other carcinogens was high. This study 
demonstrated, in part, that differences in employment opportunities may 
lead to differences in exposure and disease occurrence. 



76 



In fact, few studies of cancer survival and incidence among minority 
groups have also examined SEP. Limitations of available data are part of 
the reason for this lack: cancer is a rare disease, and a large number of 
cases is needed for such multivariate analyses; also, accurate information 
on socioeconomic variables is often not available. However, those studies 
which have examined SEP, minority status, and health have produced some 
important results. 

Dayal (21, 22) has conducted two analyses examining black-white 
differences in survival from prostate and breast cancer and the contribu- 
tion of SEP to those differences. In both studies, black-white survival 
differentials became non-significant with adjustment for SEP. A factor 
complicating the understanding of minority cancer differences is that some 
minorities present cancers at a later diagnostic stage than whites. The 
Dayal Study on breast cancer found that, even with adjustment for diagnos- 
tic stage, the black-white difference is significant. However, adjustment 
for SEP rendered the black-white survival difference non-significant. 

Lung cancer incidence rates for blacks are higher, and survival rates 
are lower than whites. A study by DeVesa and Diamond (25) reported an SEP 
gradient for lung cancer incidence in males within both black and white 
groups. Black rates were higher than white rates, and rates for low SEP 
persons in both white and black groups were poorer than rates for high SEP 
persons. The group with the lowest rates was high SEP white males, and the 
group with the highest rates was low SEP white males. However, the overall 
black-white difference lost significance when adjusted for age, area of 
residence, income, and education. Comparison of white to black males at 
the same educational level suggested that there were no significant differ- 
ences. A major shortcoming of this study was the lack of data on smoking, 
a major risk factor in lung cancer. However, the study authors felt that 
adjustment for smoking would not explain all of the differences and that 
SEP had an independent effect on lung cancer. 

Little research has been done on cancers in minority groups other than 
blacks in the United States. Some of the survival rates experienced by 
these groups are shown in the tables above. These rates are not adjusted 
for SEP, so it is not possible to determine what effect SEP may have on 
cancer incidence and survival in these groups. A study of cancer survival 
among Asians and Pacific Islanders in Hawaii reported that Caucasians had 
the lowest median survival time overall and that Chinese, who were at the 
lowest status economically, survived the longest (107). Hawaiians and 
Filipinos who were at the lowest SEP level exhibited the shortest median 
survival time. After adjustment for sex, age at diagnosis, stage of 
disease, and SEP, many of the white-minority and minority-minority differ- 
ences were non-significant. The strongest predictor of death in that study 
was stage at diagnosis, which has been associated with both SEP and 
minority group status in other studies. 

Thus there exist substantial differences between whites and minority 
groups with respect to cancer incidence and survival. Both minority status 
and SEP are associated with incidence of some cancers and with stage at 
diagnosis for many cancers. It seems apparent that minority status and SEP 
are intertwined in the etiology of and survival from cancer. 

Cardiovascular Disease 

Blacks in the United States have among the highest rates of cardiovas- 
cular disease (CHD) in the world (33, 34). Reports of CHD mortality have 



77 



shown black male rates to be higher than white rates for the past twenty 
years. Rates among black and white women are similar. Ischemic heart 
disease and stroke account for 35 per cent of mortality among blacks and 
other non-whites as a group. However, CHD incidence and mortality rates 
for other minority groups do not follow the same pattern as for blacks. A 
study of CHD mortality in Los Angeles County, California, reported blacks 
as having the highest CHD mortality; whites as second; and hispanics, 
Japanese, Chinese, and Filipinos in descending order of mortality. Black 
rates were 10 per cent higher than whites for all major cardiovascular 
diseases and 24 per cent higher for cerebrovascular diseases (32). 

Few cardiovascular disease studies have addressed the simultaneous 
issues of minority group membership and SEP. A study conducted in Evans 
County, Georgia, between 1960-1977 (34, 95) has attempted one such analy- 
sis. In this study (as in many studies), virtually all blacks were of low 
SEP. In fact, they were of lower SEP than most low status whites. This 
study reported that 20-year survival from all-cause mortality was almost 
identical for low SEP whites and blacks, and both were higher than high SEP 
whites. The risk of dying from ischemic heart disease associated with 
blood pressure, cholesterol, and smoking was similar for low status whites 
and for blacks, and both were substantially different from high status 
whites. 

A study in Charleston, South Carolina (50), comparing CHD incidence 
among black males and females, white males and females, and high SEP black 
males for the period 1961-1975 demonstrated that SEP is strongly negatively 
associated with CHD. This study reported that high SEP black males had the 
lowest incidence of all categories of CHD compared to all blacks and all 
whites, except for arteriosclerotic heart disease, for which they had the 
highest rates. Table 7 shows these results. The lower rates observed in 
high SEP black males were found for all CHD, non-fatal CHD, fatal CHD, 
acute myocardial infarction (both fatal and non-fatal), angina, and sudden 
death. The number of cases of CHD (n=12) among high SEP black males was 
low, however, and the observations are most valuable for the trend they 
suggest. 

CHD Risk Factors 

The CHD risk factors most frequently measured include hypertension, 
blood lipids, smoking, diabetes. Type A behavior, overweight, ECG abnor- 
malities, and, in some studies, heavy alcohol consumption. Blacks are 
reported (33, 34, 95) as being at greater risk for CHD from hypertension, 
diabetes mellitus, ECG abnormalities, and overweight (among black women). 
As a general pattern, these CHD risk factors operate for blacks and other 
minority groups as they do among whites. 

Several studies on the distribution of CHD risk factors have suggested 
that there is an association between SEP and risk of CHD among blacks. 
Attempts to examine the association between CHD risk factors, minority 
status, and SEP have included both ecological level and individual level 
measures. 

Research (94) by Tyroler and Cassel has reported that ecological 
measures of social disorganization are strongly associated with mortality 
from stroke among black males and females. Another study using ecological 
measures of urban stress (40) reported a positive association with systolic 
and diastolic blood pressure for blacks but not for whites. 

Kraus, Borhanl, and Franti (52), in their study of CHD risk factors, 

78 



reported a consistent negative SEP gradient for CHD risk factors and a 
consistent negative association between SEP and CHD risk within ethnic 
groups. Comparison of different minority groups at the same SEP levels 
suggests that (a) low SEP white males are at greater risk than minority 
males at the same SEP level, and (b) high SEP white males are at lower 
risk than minority males at a high SEP level, except hispanic males. For 
example, low SEP white males were at 30 per cent greater risk than low SEP 
black males, 52 per cent greater risk than low SEP Asian males, and 130 per 
cent greater risk than low SEP hispanic males. Conversely, high SEP white 
males were at 28 per cent lower risk than high SEP blacks males, 8 per cent 
lower than high SEP Asians, but 25 per cent higher risk than high SEP 
hispanics. 

A study by Stern et al. (87) on Mexican Americans reported that they 
had higher CHD risk factors levels than whites with respect to plasma 
lipids, diet, and adiposity but lower risk from cigarettes, blood pres- 
sure, and alcohol. The study made the point that Mexican Americans were 
primarily of low SEP but did not report any SEP-stratif led data. A study 
by Roberts and Lee (75) on Mexican Americans reported that adjustment for 
health practices reduced but did not completely explain the health differ- 
ence between whites and Mexican Americans. 

Data from the Hypertension Detection and Follow-up Program has been 
reported (47) showing that prevalence of hypertension among blacks and 
whites of both sexes decreases with increasing education. Furthermore, the 
difference between whites and blacks generally decreases with increasing 
education. 

The evidence presented above suggests that SEP is a powerful risk 
factor that may help to explain the higher incidence and mortality rates 
and the poorer survival rates among certain minority groups. Furthermore, 
it suggests that the association between SEP and minority status and CHD 
cannot be fully explained by adjustment for risk factors such as smoking, 
alcohol consiimption, or obesity. In short, SEP appears to exert an inde- 
pendent influence upon CHD and to partially explain the differences between 
blacks and whites. The association between SEP and CHD risk and occurrence 
among other minority groups is less clear, primarily because data on these 
groups is sparse and not generally presented with information on SEP. 

Infant Mortality 

As discussed in an earlier section, the association between infant 
mortality, low birth weight, and SEP is well established. The question to 
be examined in this section is whether the differences observed between 
blacks and whites can be explained by SEP. 

Data from 1976 on low birth weight reported by NCHS (104) show that 
(a) the percentage of infants weighing 2,500 grams or less at birth 
decreases with increasing education of mother or father, (b) that the 
percentage of low birth weight infants is greater at all education and 
income levels among blacks but declines with increasing SEP for both 
groups. Education of the mother appears to have little direct effect on 
low birth weight when prenatal care is totally absent. When prenatal care 
is present, education of the mother has a strong effect for both groups. 
The effect of prenatal care is not unrelated to SEP, however, since access 
to medical care and awareness of the need for prenatal care are probably 
both associated with SEP (89). The prevalence of low birth weight infants 
is greater among lower SEP individuals of both black and white groups and 



79 



is greater for blacks than for whites at all SEP levels. 

Infant death is strongly associated with SEP for both whites and 
blacks (103, 105) whether measured by education of father or mother or by 
family income. Blacks at all income or education levels had higher rates 
than whites. However, the black-white difference decreases as SEP 
increases. For example, the black-white difference is 12 per cent when the 
father's education is eighth grade or less and 4 per cent when the father's 
education is at high school level (105). In this data, blacks were not 
present at the highest SEP levels, so it was not possible to determine 
whether the black-white differential for infant mortality would disappear 
at higher SEP levels. These data suggest that SEP as measured by parental 
education or family income can help explain the black-white differences in 
infant mortality. Data on higher SEP blacks and other minorities are 
lacking, however. 

SEP and minority status appear to have direct effects on infant 
mortality and an indirect effect on neonatal mortality. A 1980 study by 
Brooks (12) using area measures of racial composition and Income reported 
that racial composition did not affect post neonatal mortality, whereas 
income and low birth weight together explained 65 per cent of the variance. 
Low income alone explained 57 per cent of the variance. The addition of 
racial composition to a model including low birth weight, illegitimate 
status, and low income increased the explained variance by only 1.2 per 
cent, a non-significant change, suggesting that income and not minority 
status contributed most to infant mortality. Neonatal mortality in this 
study was best explained by a model including low birthweight, low income, 
and racial composition. Low birth weight and low income were highly 
correlated (.80) in this study, as were racial composition and low income 
(.683) (57). 

Further evidence for the hypothesis that the higher rates of infant 
death and childhood experienced by blacks may be partly explained by SEP is 
provided in research by Mare (57). His research reported that for both 
blacks and whites, mother's education and family income were negatively 
associated with death for children of all ages under 19 years of age. 
Furthermore, this association increased in size with increases in the 
child's age. In general, mortality rates for white males were higher at 
most ages than rates for black males at both high and low income levels. 
Among females, the effects of income were less clear. For annual family 
income less than $10,000, black females generally had lower rates than 
white females, and this was more true at older age levels. Comparison of 
white females to black females at family incomes over $10,000 shows that 
black females suffered substantially higher mortality rates at all age 
levels. Examination of the association between mother's education and 
mortality produced somewhat different results. Mortality rates for white 
male children were lower than black male children at younger ages and 
higher at older ages for education less than twelve grades. For mother's 
education at twelve grades or more, the reverse was true, and black males 
mortality rates were lower than whites. The mortality rates for black 
females whose mothers had less than a twelfth-grade education were somewhat 
higher than white females. For those females with mothers educated at 
twelve grades or better, black females suffered substantially higher 
mortality than white females. This study demonstrated a clear association 
between SEP and childhood mortality for both blacks and whites and suggests 
that, for males at least, the higher childhood death rates suffered by 
blacks may be due to lower socioeconomic position. 



80 



These data suggest that neonatal and post-neonatal mortality are 
affected by SEP for both blacks and whites. Such socioeconomic factors as 
low income, low education, low status occupation, minority status, teenage 
pregnancy, and non-married parents are all closely associated risk factors 
in the etiology of low birth weight, neonatal mortality, and infant 
mortality. The data available on SEP was insufficient to conclude that all 
black-white differences could be explained by SEP. However, it is clear 
that SEP is a powerful risk factor in both infant and neonatal mortality 
for both blacks and whites and that increasing SEP substantially decreases 
the black-white differences with respect to low birth weight, infant 
mortality, and childhood mortality. 

Non-Disease Causes of Injury and Death: 

Minority Status and Socioeconomic Position 

There is substantial variation in injury death rates among ethnic 
groups. In general. Native Americans and blacks have the highest death 
rates from such causes, and Asian Americans have the lowest. For a number 
of non-disease causes of death, the differences between ethnic groups is 
lessened or eliminated with adjustment for some measure of SEP. This 
section reviews some available data which demonstrates these points for 
unintentional injuries, motor vehicle accidents, accidental death from 
firearms, and deaths from house fires. 

More than 160,000 Americans died in 1980 from unintentional injuries, 
including such causes as accidental ingestion of poison, poisoning by 
faulty heaters (i.e., carbon monoxide), and motor vehicle accidents (nearly 
50% of all unintentional injuries), etc. Asian Americans have the lowest 
rates, and Native Americans have the highest rates, with blacks and whites 
falling in between. All rates decline substantially with increasing 
income, although the differences between minority groups are not greatly 
reduced by such adjustment, except for Native Americans. The rate per 
100,000 for Native Americans drops by nearly 300 per cent between per 
capita income less than $3,000 and per capita income of $5,000. The rate 
for blacks drops by more than 100 per cent with adjustment for income. The 
change in rates for Asians and whites is similar to those for blacks with 
income adjustment. 

Death rates from unintentional firearm injury for whites and blacks 
are similar, with the rates for both groups declining precipitously with 
increasing income. Blacks have much higher death rates from housefires 
than whites. However, this difference declines substantially with adjust- 
ment for income. The difference between blacks and whites at per capita 
area incomes less than $3,000 is three-fold. At incomes greater than 
$6,000, the difference is less than 100 per cent higher. The black-white 
death rate difference for occupants of motor vehicles and for pedestrian 
deaths declines with income adjustment also, although blacks have lower 
rates than white from the former cause and higher rates than whites from 
the latter cause. In general, it may be said that differences in non- 
disease mortality rates between whites and minorities, especially blacks 
and Native Americans, are diminished with adjustment for income (4). 

Part 2 

In the preceding sections, we have argued that socioeconomic position 



81 



ought to be considered as a potential explanatory variable when considering 
minority and white health. We have reviewed the evidence that SEP is 
consistently related to a variety of health outcomes for different ages, 
places, and times. We have briefly presented evidence of the strong asso- 
ciation between SEP and membership in minority groups and have reviewed 
much of the available evidence that differences in the distribution of SEP 
may account for the differential health experience of whites and minori- 
ties. Our intent was to make the argument plausible. The evidence which 
we have presented, in our opinion, supports such an argument. However, it 
is important to specify why low SEP is associated with poor health. 

It has has been argued that such associations reflect the downward 
drift of less healthy individuals into lower socioeconomic strata. However, 
there are a number of reasons to believe that this is not what accounts for 
the association between SEP and health. Although it is undoubtedly true 
that long-term illness has an impact on income, it is difficult to see how 
such an explanation might apply to groups of individuals. Given the over- 
all pattern of lower SEP associated with minority status, it is hard to 
argue that this lower SEP is the result of poorer health. Indeed, in one 
analysis (18), income differences between minorities and whites were 
substantially reduced when there was statistical adjustment for age, educa- 
tion, occupational prestige, hours worked in the previous week, and average 
income of the state of residence. This adjustment accounted for 57 per 
cent of the income differences between whites and blacks. The comparable 
figures for Mexican Americans, Puerto Ricans, and American Indians were 49 
per cent, 93 per cent, and 70 per cent, respectively. In short, the lower 
SEP of minorities is not due to poorer health, rather it reflects an 
overall pattern of disadvantage. 

The argument is also not plausible given the variety of measures of 
SEP shown to be associated with poorer health. As we have pointed out 
earlier, although each of the measures of SEP has some interpretive 
problems, the overall pattern across measures is sufficiently consistent to 
be compelling. 

Differential patterns of risk factors are often proposed as explana- 
tions for SEP gradients of disease. Our review has not turned up 
consistent patterns of risk factor differences which could account for the 
disparities between minority and white health. There are few studies which 
allow us to examine in detail the validity of these explanations. The few 
studies there are suggest that such explanations do not adequately account 
for SEP gradients. With respect to cardiovascular disease, there are three 
studies which have had the opportunity to directly address this issue. In 
one study of cardiovascular disease among 18,000 male British civil 
servants, it was possible to examine the contribution of serum cholesterol, 
smoking, hypertension, and other cardiovascular risk factors to the gradi- 
ent of cardiovascular disease associated with SEP, measured by broad 
occupational groupings (76). In these analyses, there was a consistent 
gradient of cardiovascular mortality associated with SEP; those in adminis- 
trative classifications had the lowest rates, followed by those in 
professional/executive positions, clerical positions, and the remainder. 
Figure 2 presents the results from this study when cardiovascular risk 
factors were introduced. Taking into account the standard risk factors for 
cardiovascular disease did not alter the gradient associated with SEP. 
Similar results were found by Salonen (80) in Finland, and Holme et al. in 
Oslo (46). 



82 



Turning to lung cancer, SEP gradients do not seem to be entirely due 
to higher rates of smoking among lower SEP groups. Although lower SEP 
groups such as blacks may have higher rates of current smoking, some 
evidence suggests that they smoke fewer cigarettes and tend to use weaker 
tobacco products (86, 111). Analyses of data from the Third National 
Cancer Study (25), the Washington County, Maryland, Study (15), and other 
studies (109) suggest that adjustment for level of smoking does not elimi- 
nate the SEP gradients for lung cancer incidence. 

Differences in access to medical care are also often proposed as an 
explanation for SEP gradients in health. However, such factors do not 
adequately account for SEP gradients. The presence of the National Health 
Service in England and Wales and the equivalent services in the 
Scandinavian countries would seem to provide reasonable access to care. 
However, in England and Wales, Sweden, and Finland, there are substantial 
SEP gradients of health. The evidence in England and Wales is that these 
gradients did not change substantially following introduction of the 
National Health Service (45). Similarly, the last 20 years in the United 
States have seen large changes in the accessibility of medical care to the 
poor. Between 1964 and 1976, persons in the lowest fifth of the income 
distribution increased utilization of physician and hospital services by 
one third (54, 70). Similar changes in health insurance coverage have 
occurred, particularly for the aged. However, despite these changes, 
national data do not indicate any major changes in the SEP gradient of 
prevalence or mortality. This is not to say that such changes have not had 
important health consequences but only that they do not seem to have 
resulted in major changes in the association between SEP and health. 

Further evidence that differences in levels of risk factors or medical 
care do not account for SEP gradients of health comes from analyses we have 
recently completed at the Human Population Laboratory in Alameda County, 
California (38). In these analyses, we examined the nine-year mortality 
experience of a representative sample of adults in Oakland, California, 
beginning in 1965. At that time, a portion of Oakland was federally desig- 
nated as a poverty area, based on rates of unemployment and income reported 
in the 1960 Census. Table 8 shows some of the characteristics of the 
poverty area compared to the nonpoverty area (44). Approximately 41 per 
cent of Oakland's population lived in the designated poverty area. The 
poverty area exhibited disproportionate levels of unemployment for both men 
and women, poorer health measured in a variety of ways, and poorer quality 
of housing. Those in the poverty area had three times the rate of unem- 
plojrment, twice the number with an eighth grade education or less, two and 
one-half times the rate of inadequate incomes, and almost two and one-half 
times higher rates of no health insurance compared to residents of the 
nonpoverty area. 

We were interested in the extent to which this pervasive pattern of 
socioeconomic disadvantage would be associated with poorer health among the 
residents of the poverty area. Furthermore, because data were available 
for each participant, we were able to ascertain if poorer health among the 
poverty area residents might be due to differences in age, income, baseline 
health status, lack of medical care, minority group status, health prac- 
tices such as smoking and alcohol consumption, or psychological factors 
such as depression. 

When we examined the nine-year mortality experience of this cohort, 
residents of the poverty area were at significantly increased risk of 
death. Furthermore, when all of the above factors were taken into account 
statistically, poverty area residents had 46 per cent higher mortality from 



83 



all causes. In other analyses (49), we have shown that this survival 
disadvantage persists over 17 years of follow-up. In addition, when 
adjustment for residence in the poverty or nonpoverty area was carried out, 
there were no significant differences in mortality for whites and non- 
whites. 

These results suggest that we need a broader based approach to our 
examination of SEP gradients in health and their value in explaining 
minority health experience. Poverty areas are characterized by a large 
number of vectors of disadvantage ranging from poorer environmental 
quality, higher unemplojnnent , lower income and education, higher rates of 
crime, greater social isolation, poorer services, to higher levels of 
reported stress. It is of great significance that these are the areas in 
which a disproportionate number of minority group members live. 

This clustering of high socio-environmental demands such as pollution, 
bad housing, and crime, coupled with low resources such as low income, 
social isolation, and inadequate services, may be what is responsible for 
SEP gradients of health. Several research efforts, using ecological 
measures of social area characteristics, have produced results relevant to 
this approach. Jenkins et al. (48) found census tract SEP indicators such 
as low occupational status, substandard housing, and low median education 
to be associated with mortality from hypertensive diseases. They also 
found significant associations for mortality due to all respiratory 
diseases, cerebrovascular disease (excluding hypertension), and ischemic 
heart disease. Dayal et al. (20) has reported that residence in low socio- 
economic level neighborhoods is associated with mortality from both lung 
and non-lung cancers, suggesting that both air pollution and socioeconomic 
variables are associated with poorer health among low SEP groups. This 
association was not affected by adjustment for race. Harburg et al. (40), 
using an area measure of social stress, found a significant cross- 
sectional association between systolic blood pressure and residence in such 
areas for black males and females. Similarly, area measures of social 
disorganization were found by Tyroler and Cassel (94) to be positively 
associated with stroke mortality. Finally, a step toward integrating 
ecologic and individual level variables has been taken by Hakama et al. 
(39) in an analysis of cancer of the breast and cervix. The findings in 
his study suggest that social and physical environmental factors might be 
relatively more important in the etiology of breast cancer than cervical 
cancer. 

Conclusions 

Studies which have examined minority/white differentials in health 
have often alluded to differences in culture, lifestyle, or genetics and 
have generally ignored the role of SEP. However, minority status and SEP 
are closely associated, and the evidence suggests that a portion of the 
difference in health between whites and minorities can be explained by 
differences in SEP. Furthermore, SEP gradients of health cannot, in many 
cases, be explained by differences in risk factor levels or differences in 
medical care. Finally, in analyses of all-cause mortality, survival 
differences in cancer of the breast and prostate, male lung cancer 
incidence, and mortality from coronary heart disease, minority /white 
differentials in health decrease significantly when SEP is taken into 
account. For many other outcomes, the evidence suggests a diminution of 
minority /white differentials with adjustment for SEP. 



84 



These results suggest that it is not minority status, itself, which 
leads to poorer health. Indeed, some minority groups evidence, for some 
outcomes, better health. Rather, it is the association of low SEP with 
minority group membership which has consequences for health. 

It is clear from this review that more research and analysis is needed 
on the health status of minority groups. Much of the available data only 
focuses upon white versus black differences and excludes other minority 
groups or includes them in a non-white grouping. As we have demonstrated 
in the preceding sections, there are significant differences between the 
various minority groups with respect to both SEP and to health. However, 
our understanding of the role of SEP in minority health is compromised by 
the lack of data on patterns of incidence, survival, and medical care 
utilization. As has become apparent in the consideration of the declines 
in CHD mortality, such information may be critical to our understanding of 
mortality differences. These data may be particularly significant in 
unraveling the impact of SEP on minority and white health. 

Similarly, the effort to understand minority health experience would 
be greatly improved by analyses that also examine the role of SEP. The 
evidence presented in this report strongly suggests that such analyses 
would be particularly helpful in clarifying the reasons for the substantial 
differences between whites and minorities observed for most major disease 
outcomes and all-cause mortality. 

As we have discussed previously, the measurement of SEP is prob- 
lematic. The most commonly used measures — income, occupation, and 
education — may not adequately assess the effects of SEP on health. For 
example, a white collar worker and a blue collar worker may have the same 
income and education but experience a different social and physical 
environment at work. Similarly, a highly educated person may have a rela- 
tively low income. Also, different measures may affect health in different 
ways. For example, income may affect health through the ability to 
purchase adequate medical care, while occupation may affect health through 
differential social and physical exposures on the job. Finally, one or 
two-time measures of SEP may fail to capture the lifetime exposures that 
individuals actually incur. As we discussed earlier, much research 
suggests that social and physical risk factors may co-occur in consistent 
patterns which are not random but are determined by larger socioeconomic 
forces. Our understanding of the role of SEP in minority health would be 
enhanced by examination of both ecological and individual-level risk 
factors. The studies by Jenkins et al., Dayal et al., Tyroler and Cassel, 
and Harburg et al. and Hakama et al. (20, 39, 40, 48, 94) indicate that SEP 
involves more than measures of income, education, and occupation can 
capture. As we have amply demonstrated, a large proportion of minority 
group members are also low SEP group members. Therefore, our understanding 
of minority health will be improved if analyses capturing the complex 
interrelationships between these different levels and types of risk factors 
can be attempted. An approach which combines environmental and individual 
level analyses can provide a method for a more coherent description of 
disease etiology than approaches which focus on only one level of analysis. 
This approach could be especially important in the investigation of 
minority health and SEP, factors which are multi-faceted and which exert 
their effects at both group and individual levels. Without such an 
approach, it is unlikely that we will be able to understand the reasons for 
the differential health experience of minorities and whites. 



85 



TABLE 1 

Income by Minority Status for Currently Employed Persons 

17 Years of Age and Older (per cent), 1976 



Income 



Hispanic 



Black 



All other 
(including white) 



$5,000 

$5,000 to 
$9,999 

$10,000 to 
$14,999 

$15,000+ 



13.5 
30.5 

26.9 

29.1 



17.2 
30.6 

28.0 

28.0 



7.4 
17.6 

24.5 

50.5 



Totals 



3,662 



7,418 



69,463 (number) 



Source: (23) 



86 



TABLE 2 
Occupational Distribution of Minority Groups 
Ratio of White to Minority 



Category 



Group 



Black (112) Asian (88) 



American 
Indian (88) 



Hispanics (23) 



Sex 



M 



M 



M 



M 



White 
collar 

Blue 
collar 

Farm 

Service 

Employed 



1.39 1.59 1.03 



.77 



,85 



,96 1.50 2.04 



.88 1.42 



,68 



.77 



1.56 1.08 1.20 1.03 .78 .66 

.56 .49 1.03 .35 .57 .57 

1.08 1.07 1.90 1.39 2.95 1.82 



4.70 



1.54 



,13 



Source: (23) 



87 



TABLE 3 
Educational Attainment by Sex and Race, 1978 



Females 



% 
White 



% 
Black 



Males 



% 
White 



% 
Black 



to 8th 
grade 

4 years of 
college 



4.3 



21.4 



4.0 



12.6 



4.3 



27.6 



6.5 



10.7 



Source: (97) 



88 



TABLE 4 
Median Years of School by Minority Status, 1970 





American 










Indian 


Japanese 


Chinese 


Filipino 


Women 


9.9 


12.4 


12.2 


12.7 


Men 


9.7 


12.6 


12.5 


11.6 



Source: (97) 



89 



TABLE 5 
Age-Adjusted Rates for Selected Conditions by Income and Minority Group 



Condition 



Hispanlcs 



Blacks 



All Others 



$5,000 $15,000+ $5,000 $15,000+ $5,000 $15,000+ 



Limit of 19.7 

activity 

due to chronic 

condition 

Hospitalization 11.7 
in short-term 
stay hospital 

Days of 16.3 

restricted 

activity 



12.2 



9.1 



4.2 



24.9 



13.7 



12.8 



10.4 



9.0 



8.5 



23.0 



12.6 



10.1 



10.8 



9.7 



5.5 



Source: (106) 



90 



TABLE 6A 
Ratio of White/Minority Group 5-Year Survival Rates for Males 



Minority 
Group 




All Sites 


Lung 


Cancer Site 
Stomach 


Rectum 


Prostate 


Anglo 




1.00 


1.00 


1.00 


1.00 


1.00 


Hispanic 




.97 


1.00 


.60 


1.09 


.92 


Black 




1.07 


1.14 


.82 


1.40 


1.10 


American 


Indian 


1.48 


2.00 


1.50 


1.67 


1.55 


Chinese 




1.15 


.67 


1.13 


.81 


.94 


Japanese 




.91 


.80 


.43 


.78 


.82 


Filipino 




1.11 


.89 


.82 


1.03 


.80 


Hawaiian 




1.29 


.80 


.75 


.75 


.98 


TABLE 6B 








Breast 




Corpus Uteri 


Anglo 






1.00 






1.00 


Hispanic 






.99 






.99 


Black 






1.16 






1.65 


American 


Indian 




1.25 






1.23 


Chinese 






.90 






.96 


Japanese 






.79 






.95 


Filipino 






.97 






.95 


Hawaiian 






.96 






1.15 



Source: (113) 



91 



TABLE 7 

Age-Adjusted Incidence of Cardiovascular Disease by Sex and 

Minority Status, and SEP 



Males 
Blacks Whites Hi-SEP Black 



Females 
Blacks Whites 



All CHD 
CHD Deaths 



131.7 
79.8 



188.4 
93.8 



61.2 
38.3 



161.0 
62.2 



113.8 
46.3 



Source: (95) 



92 



TABLE 8 
Poverty Area Characteristics 



41% of Oakland's Population 



66% of unemployed males 14 years or older 

61% of unemployed females 14 years or older 

85% of Oakland General Assistance recipients 

79% of AFDC recipients 

79% of aid to disabled 

63% of blind receiving aid 

65% of police work load 

68% of active TB cases 

69% of Oakland's deteriorating housing units 

75% of Oakland's non-owner occupied units 

89% of Oakland's housing units with shared or no bathroom 



Source: (44) 



93 



Figure 1 

Age and Sex-Adjusted Survival for 

Blacks and Whites in the Alameda County Study 

In la, blacks have significantly poorer survival (p < 0.004); when there is 

adjustment for SEP (lb), this difference is no longer significant (p > 0.05). 



la 


Survival Curves by Race Adjusted for Age and Sex 


Black versus White 


1.00 


p. 




>^ -a 


0.95 




Probability 

o 

O 


\ 6, 

\ o White 


1 


Black \ ~<3. 


Estimated Sui 

P o 




0.75 


— 1 1 r 1 1 1 1 1 1 


2 4 6 8 10 12 14 16 18 


Years in Study 



lb 


Survival Curves by Race Adjusted for Age, Sex, 


and Income 


Black versus White 


1.00 


'V^ 


0.95 




Probability 

o 
o 


'V -Q^ White 


? 


Black\ '^•b 


Estimated Survi 

o 

oo 




0.80 




0.75 


— II 


( 


) 2 4 6 8 10 12 14 16 18 


Years in Study 



94 



Figure 2 

Relative Risk of Death from 

CHD Compared to Administrative Classification 

Adapted with permission from reference 76. 



Relative Risk 
(Log Scale) 

4.0-1 



3.0- 



2.0' 



1.5- 



1.0 





Cholesterol 
Smoking 
li: '.y Blood Pres. 
Others 
Unexplained 





D 



1.0 







Administrative Professional/ 

Executive 



3.2 



Clerical 



4.0 



2.6 




Others 



95 



REFERENCES 



1. Antonovsky A. Social class, life expectancy, and overall mortality. 
Milbank Mem Fund Q 1967; 45:31-73. 

2. Antonovsky A, Bernstein J. Social class and infant mortality. Soc 
Sci Med 1977; 11:453-70. 

3. Austin DF, King M, Roe K. Cancer incidence in white women of the 
San Francisco Bay Area. National Cancer Institute Monograph, 1979; 
53:95-101. 

4. Baker SP, O'Neill B, Karpf RS. The injury fact book. Lexington, MA: 
DC Heath, 1984. 

5. Behm H. Socioeconomic determinants of mortality in Latin America. 
In: Proceedings of the meeting on socioeconomic determinants and 
consequences of mortality, WHO, Geneva, 1980. 

6. Berkman LF, Breslow L. Health and ways of living: the Alameda 
County Study. Cambridge, MA: Oxford University Press, 1983. 

7. Black D, Morris JN, Smith C, Townsend P. Inequalities in health: 
the Black Report. Middlesex, UK: Penguin Books, 1982. 

8. Blaxter M. Social class and health inequalities. In: Carter CO, 
Peel J, eds. Equalities and inequalities in health. London: 
Academic Press, 1976:111-125. 

9. Bradshaw J, Lawton D, Staden F, Weale J, Weekes A. Area variations 
in infant mortality, 1975-77. J Epidemiol Comm Health, 1982; 36:11- 
16. 

10. Brennan ME, Lancashire R. Association of childhood mortality with 
housing status and unemplojnnent . J Epidemiol Comm Health, 1978; 32: 
28-33. 

11. Bright M. A follow-up study of the Commission on Chronic Illness 
Morbidity Survey in Baltimore. II. Race and sex differences in 
mortality. J Chron Dis 1967; 20:717-29. 

12. Brooks CH. Social, economic and biologic correlates of infant mor- 
tality in city neighborhoods. J Health Soc Behav 1980; 21:2-11. 

13. Caldwell JC. Education as a factor in mortality decline: an exami- 
nation of Nigerian data. In: ed. Proceedings of the meeting on 
socioeconomic determinants and consequences of mortality. WHO, 
Geneva, 1980, 

14. Chase HC, The relationship of demographic factors and medical care 
to adverse pregnancy outcomes. In: Kelly S, Hook EB, Janerich DT, 
Porter IH, eds. Birth defects - risks and consequences. New York: 
Academic Press, 1976:139-57. 

96 



15. Comstock GW, Tonascla JA. Education and mortality in Washington 
County, Maryland. J Health Soc Behav 1978; 18:54-61. 

16. Congressional Research and Service. Infant mortality, a report pre- 
pared by the Congressional Research Service for the Subcommittee on 
Health and the Environment and the Subcommittee on Oversight and 
Investigations of the Committee on Energy and Commerce, US House of 
Representatives. Washington, DC: US Government Printing Office, 
1984. 

17. Cooke KR, Skegg DCG, Eraser J. Socio-economic status, indoor and 
outdoor work, and malignant melanoma. Int J Can 1984; 34:57-62. 

18. Cooper R, Steinhauer M, Miller W, David R, Schatzkin, A. Racism, 
society and disease: an exploration of the social and biological 
mechanisms of differential mortality. Int J Health Serv 1981; 
11(3):39-414. 

19. Cunningham LS, Kelsey JL. Epidemiology of musculoskeletal 
impairments and associated disability. Am J Public Health 1984; 
74:574-79. 

20. Dayal H, Chiu CY, Sharrar R, Mangen J, Rosenwaike I, Shapiro S, 
Henley AJ, Goldberg-Alberts R. Ecologic correlates of cancer mor- 
tality patterns in an industrialized urban population. J Nat Cancer 
Soc 1984; 73:565-74. 

21. Dayal HH, Chiu C. Factors associated with racial differences in 
survival for prostatic carcinoma. J Chron Dis 1982; 35:553-60. 

22. Dayal HH, Power RN, Chiu C. Race and socioeconomic status in survi- 
val from breast cancer. J Chron Dis 1982; 35(8) :675-83. 

23. Department of Commerce. A statistical portrait of women in the 
United States: 1978 special studies. Series P-23, No. 100. 
Washington, DC: US Government Printing Office, 1980. 

24. Derrienic F, Ducimetiere P, Kritsikis S. La mortalite cardiaque des 
Francais actif d'age moyen selon leur categorie socio-professionelle 
et leur region de domicile. Revue d'Epidemiologie, Medecine Sociale, 
at Sante Publique 1977; 25:131. 

25. DeVesa SA, Diamond EL. Socioeconomic and racial differences in lung 
cancer incidence. Am J Epidemiol 1983; 118:818-31. 

26. Dorm HF, Cutler SJ. Morbidity from cancer in the United States. 
Public Health Monograph No. 56. Washington, DC: US Government 
Printing Office, 1959. 

27. Ericson A, Eriksson M, Westerholm P, Zetterstrom R. Pregnancy out- 
come and social indicators in Sweden. Acta Paediatrica Scandinavica 
1984; 73:69-74. 

28. Feldman JG, Stein SC, Klein RJ, Kohl S, Casey G. The prevalence of 



97 



neural tube defects among ethnic groups in Brooklyn, New York. J 
Chronic Diseases 1982; 35(l):53-60. 

29. Fisher S. Relationship of mortality to socioeconomic status and some 
other factors in Sydney in 1971. J Epidemiol Coram Health 1978; 
32:41-46. 

30. Forsdahl A. Are poor living conditions in childhood and adolescence 
an important risk factor for arteriosclerotic heart disease. Br J 
Prev Soc Med 1977; 31:91-5. 

31. Frey RS. The socioeconomic distribution of mortality rates in 
Des Moines, Iowa. Public Health Rep 1982; 97:545-49. 

32. Frils R, Nanjundappa G, Prendergast TJ, Welsh M. Coronary heart 
disease mortality and risk among hispanlcs and non-hispanics in 
Orange County, California. Public Health Rep 1981; 96(5) :418-20. 

33. Gillum RF. Coronary heart disease in black populations. I. 
Mortality and morbidity. Am Heart J 1982; 104:839-851. 

34. Gillum RF, Liu KC. Coronary heart disease mortality in United States 
blacks, 1940-1978. Trends and unanswered questions. Am Heart J, 
1984; 108:728-32. 

35. Goldstein H. Factors related to birth weight and perinatal mor- 
tality. Br Med Bull 1981; 37(3) :259-64. 

36. Gortmaker SL. Poverty and Infant mortality in the United States. Am 
Soc Rev 1979; 44:280-97. 

37. Gorwitz K, Dennis R. On the decrease in the life expectancy of black 
males In Michigan. Public Health Rep 1976; 91:141-45. 

38. Haan MN, Kaplan GA, Camacho-Dickey T. Poverty and health: a pros- 
pective study of Alameda County residents. Presented at 17th Annual 
Meetings of the Society for Epidemiologic Research, Houston, TX, 14 
June 1984. 

39. Hakama M, Hakulinen T, Pukkala E, Saxen E, Teppo L. Risk indicators 
of breast and cervical cancer on ecologic and individual levels. Am 
J Epidemiol 1976; 116(6) :990-1000. 

40. Harburg E, Erfurt JC, Chape C, Hauenstein LS, Schull WJ, Schork MA. 
Socloecological stressor areas and black-white blood pressure: 
Detroit. J Chron Dis 1973; 26:595-611. 

41. HDFP Cooperative Group. Race, education and prevalence of hyperten- 
sion. Am J Epidemiol 1977; 106(5) :351-361 . 

42. Hlrayama T. An epidemiologic study of oral and pharyngeal cancer in 
Central and Southeast Asia. Bulletin WHO; 1969; 34:41-69. 

43. Hlrayama T. Epidemiology of stomach cancer. Gann 1971; 11:3-19. 



98 



44. Hochstim JR, Athanasopoulos DA, Larkins JH. Poverty area under the 
microscope. Am J Public Health 1968; 58(10) : 1815-27 . 

45. Hollingsworth JR. Inequality in levels of health in England and 
Wales, 1891-1971. J Health Soc Behav 1981; 22:268-83. 

46. Holme I, Helgeland A, Hermann I, Leren P, Lund-Larsen G. Four-year 
mortality by some socioeconomic indicators: the Oslo Study. J 
Epidemiol Comm Health 1980; 34:48-52. 

47. Igra A, Stavig RG, Leonard AR. Hypertension and related health 
problems in California: results from the 1979 California 
Hypertension Survey. California Department of Health Services, 
August, 1982. 

48. Jenkins CD, Tuthill RW, Tannenbaum SI, Kirby C. Social stressors and 
excess mortality from hypertensive diseases. J Human Stress 1979; 5: 
29-40. 

49. Kaplan GA, Seeman TE, Cohen RD, Guralnik J, Knudsen LP. Mortality, 
morbidity, and functional ability in an aging population: evidence 
from the Alameda County Study. Presented at the American Public 
Health Association Annual Meeting, Anaheim, CA, 14 November 1984. 

50. Keil JE, Loadholt CB, Weinrich MC, Sandifer SH, Boyle E. Incidence 
of coronary heart disease in blacks in Charleston South Carolina. Am 
Heart J 1984; 108:779-86. 

51. Kitagawa EM, Hauser PM. Differential mortality in the United States: 
a study in socioeconomic epidemiology. Cambridge, MA: Harvard 
University Press, 1973. 

52. Kraus JF, Borhani NO, Franti CE. Socioeconomic status, ethnicity and 
risk of coronary heart disease. Am J Epidemiol 1980; 111(4) :407-14. 

53. Kuller LH, Cooper M, Perper J, Fisher R. Myocardial infarction and 
sudden death in an urban community. Bull NY Acad Sci 1973; 49:1973. 

54. Lerner M, Stutz RN. Have we narrowed the gap between the poor and 
the non-poor? Part II. Narrowing the gaps, 1959-1961 to 1969-1971: 
mortality. Med Care 1977; 15:620-35. 

55. Lipworth L, Abelin T, Connelly RR. Socio-economic factors in the 
prognosis of cancer patients. J Chron Dis 1970; 23:105-16. 

56. Logan WPD. Cancer mortality by occupation and social class, 1851- 
1971. Office of Population Censuses and Surveys-Studies on Medical 
and Population Subjects, 1982; No. 44:1-253. 

57. Mare RD. Socioeconomic effects on child mortality in the 
United States. Am J Public Health 1982; 72:539-47. 

58. Marmot MG, Adelstein AM, Robinson N, et al. Changing social class 
distribution of heart disease. Br Med J 1978; 2:1109-12. 



99 



59. Martinez I. Cancer of the esophagus in Puerto Rico. Mortality and 
Incidence analysis, 1950-1964. Cancer 1964; 17:1279-88. 

60. McCormick MC, Shapiro S, Starfield B. High risk young mothers: 
infant mortality and morbidity in four areas in the United States, 
1973-1978. Am J Public Health 1984; 74(1):18-21. 

61. Michaels D. Occupational cancer in the black population: the health 
effects of job discrimination. J Nat Med Assn 1983; 75(10) :1014- 
1018. 

62. Millar WJ. Sex differentials by income level in urban Canada. 
Canadian J Public Health 1983; 74:329-34. 

63. Morgan M, Chinn S. ACORN group, social class and child health. J 
Epidemiology Community Health 1983; 37(3) :196-203. 

64. Morgenstern H. The changing association between social status and 
coronary heart disease in a rural population. Soc Sci Med 1980; 14A: 
191-201. 

65. Morgenstern H. Uses of ecologic analysis in epidemiologic research. 
Am J Public Health 1982; 72:1336-44. 

66. Morgenstern H. Socioeconomic factors: concepts, measurements. In: 
Measuring psychosocial variables in epidemiological studies of 
cardiovascular disease. Galveston, TX: NHLBI, in press. 

67. Morris JN. Social inequalities undiminished. Lancet 1979; 1(8107): 
87-90. 

68. Mustacchi P, Millmore D. Racial and occupational variations in 
cancer of the testis: San Francisco, 1956-1965. J Nat Can Inst 
1976; 56:717-33. 

69. Nayha S. Social group and mortality in Finland. Br J Prev Soc Med 
1977; 31:231-7. 

70. Newacheck PW, Butler LH, Harper AK, Piontkowski DL, Franks PE. 
Income and illness. Med Care 1980; 18:1165-76. 

71. Nomura A, Kolonel L, Rellahan W, Lee J, Wegner E. Racial survival 
patterns for lung cancer in Hawaii. Cancer 1981; 48:1265-71. 

72. Page WF, Kuntz AJ. Racial and socioeconomic factors in cancer 
survival: a comparison of Veterans Administration results with 
selected studies. Cancer 1980; 45:1029-40. 

73. Paneth N, Wallenstein S, Kiely JL, Susser M, Social class indicators 
and mortality in low birth weight infants. Am J Epidemiol 1982; 
116(2):364-75. 

74. Pearce NE, Davis PB, Smith AH, Foster FH. Mortality and social class 
in New Zealand. II: Male mortality by major disease groupings. 



100 



New Zealand Med J 1983; 96:711-6. 

75. Roberts RE, Lee ES. The health of Mexican Americans: evidence from 
the Human Population Laboratory studies. Am J Public Health 1980; 
70:375-84. 

76. Rose G, Marmot MG. Social class and coronary heart disease. Br 
Heart J 1981; 45-13-9. 

77. Ross RK, McCurtls JW, Henderson BF, et al. Descriptive epidemiology 
of testicular and prostatic cancer in Los Angeles. Br J Cancer 1979; 
39-284-92. 

78. Rudov MH, Santangelo N. Health status of minorities and low income 
groups. USDHEW, Public Health Service, 1979. 

79. Rush D, Cassano P. Relationship of cigarette smoking and social 
class to birth weight and perinatal mortality among all births in 
Britain, 5-11 April 1970. J Epidemiol Comm Health 1983; 37:249-55. 

80. Salonen JT. Socioeconomic status and risk of cancer, cerebral 
stroke, and death due to coronary heart disease and any disease: a 
longitudinal study in eastern Finland. J Epidemiol Comm Health 1982; 
36(4):294-7. 

81. Seldman H. Cancer death' rates by site and sex for religious and 
socioeconomic groups in New York City. Environ Res 1970; 3:234-50. 

82. Shapiro S, Weinblatt E, Frank CW, Sager RV. Social factors in the 
prognosis of men following first myocardial infarction. Milbank Mem 
Fund Q 1970; 48:37-50. 

83. Simpson SP. Variation in infant mortality rates among census tracts 
in Hawaii. Hawaii Med J 1983; 414-16. 

84. Simpson SP. Causal analysis of infant deaths in Hawaii. Am J 
Epidemiol 1984; 119:1024-9. 

85. Stanley MG, Shaw H, McCabe K. Effects of socioeconomic status and 
race on weight-defined and gestational prematurity in the United 
States. In: Reed D, Stanley F, eds. Epidemiology of prematurity. 
Urban and Schwarzenberg, 1977. 

86. Sterling TD, Welnkam JJ. Smoking patterns by occupation, industry, 
sex, and race. Arch Environ Health 1978; 33:313-7. 

87. Stern MP, Haskell WL, Wood PDS, Osann KE, King AB, Farquhar JW. 
Affluence and cardiovascular risk factors in Mexican Americans and 
other whites in three northern California communities. J Chron Dis 
1975; 28:623-36. 

88. Stockwell EG, Wicks JW, Adamchak DJ. Research needed on 
socioeconomic differentials in US mortality. Public Health Rep 1978; 
93:666-72. 



101 



89. Sunderland R. Dying young in traffic. Arch Dis Childhood 1984; 59: 
754-7. 

90. Syme SL, Berkman LF. Social class, susceptibility and sickness. Am 
J Epidemiol 1976; 104:1-8. 

91. Syme SL, Cakes TW, Friedman GD, Feldman R, Siegelaub AB, Collen M. 
Social class and racial differences in blood pressure. Am J Public 
Health 1974; 64(6) :619-20. 

92. Taylor EM, Emery JL. Family and community factors associated with 
infant deaths that might be preventable. Br Med J 1983; 287:871-4. 

93. Torgersen 0, Petersen, M. The epidemiology of gastric cancer in 
Oslo: cartographic analysis of census tracts and mortality rates of 
sub-standard housing areas. Br J Cancer 1956; 10:299-306. 

94. Tyroler HA, Cassel J. Health consequences of culture change II. 
The effect of urbanization on coronary heart mortality in rural 
residents. J Chron Dis 1964; 17:167-77. 

95. Tyroler HA, Knowles MG, Wing SB, Logue EE, Davis CE, Heiss G, Heyden 
S, Hames CG. Ischemic heart disease risk factors and twenty-year 
mortality in middle-age Evans County blacks males. Am Heart J 1984; 
108:738-46. 

96. USDHHS. Mortality from diseases associated with smoking: 
United States, 1960-1977. Vital Health Stat; Series 20, No. 7. 

97. US Dept. of Commerce and Census. Social indicators. III. Selected 
data on social conditions and trends in the United States. US Dept. 
of Commerce, Dececember 1980. 

98. USDHEW. Prevalence of chronic conditions of the genito-urinary, 
nervous, endocrine, metabolic, and blood and blood-forming systems 
and of other selected conditions. Vital Health Stat; Series 10, No. 
109. 

99. USDHEW. Prevalence of chronic skin and musculoskeletal conditions, 
United States, 1976. Vital Health Stat; Series 10, No. 124. 

100. USDHEW. Disability days. United States, 1980. Vital Health Stat 
1983; Series 10, No. 143. 

101. USDHEW. Prevalence of selected chronic circulatory conditions. 
United States, 1972. Vital Health Stat; Series 10, No. 94. 

102. USDHEW. A study of infant mortality from linked records by birth 
weight, period of gestation and other variables. United States, 1960, 
live birth cohort. Vital Health Stat 1972; Series 20, No. 12. 

103. USDHEW. Comparison of neonatal mortality from two cohort studies: a 
study of infant mortality from linked records. Vital Health Stat 
1972; Series 20, No. 13. 



102 



104. USDHEW. Factors associated with low birth weight. United States, 
1976. Vital Health Stat 1976; Series 21, No. 37. 

105. USDHEW. Infant mortality rates: socioeconomic factors. Vital 
Health Stat 1972; Series 22, No. 14. 

106. USDHEW. Health characteristics of minority groups. United States, 
1976. Advance Data 1976, Kept. No. 27. 

107. Wegner EL, Kolonel LN, Nomura AMY, Lee J. Racial and socioeconomic 
status differences in survival of colorectal cancer patients in 
Hawaii. Cancer 1982; 49:2208-16. 

108. WHO. Proceedings of the meeting on socioeconomic determinants and 
consequences of mortality. Geneva: WHO, 1980. 

109. William RR, Horm JW. Association of cancer sites with tobacco and 
alcohol consumption and socioeconomic status of patients: interview 
study from the Third National Cancer Study. J Nat Cancer Inst 1977; 
58:525-47. 

110. Wynder EL, Bross IJ. A study of etiological factors in cancers of 
the esophagus. Cancer 1961; 14:389-413. 

111. Wynder EL, Stellman, SD. Comparative epidemiology of tobacco related 
cancers. Cancer Res 1977; 37-4608-22. 

112. Yeracaris CA, Kim JH. Socioeconomic differentials in selected causes 
of death. Am J Public Health 1978; 68:432-51. 

113. Young JH, Ries LG, Pollack ES. Cancer patient survival among ethnic 
groups in the United States. J Nat Cancer Inst 1984; 73(2) :341-52. 



103 



Associations of Health 
Problems with Ethnic 
Groups as Reflected in 
Ambulatory Care Visits 



M. Alfred Haynes, M.D. 

President/ Dean 

Charles R. Drew Postgraduate Medical School 

Los Angeles, California 

Girma Wolde-Tsadik, Ph.D. 

Associate Professor 

Department of Internal Medicine 

Charles R. Drew Postgraduate Medical School 

Los Angeles, California 

Paul Juarez, Ph.D. 

Health Planner 

Center for Preventive and Community Medicine 
Charles R. Drew Postgraduate Medical School 
Los Angeles, California 



ASSOCIATIONS OF HEALTH PROBLEMS WITH ETHNIC GROUPS 

IN AMBULATORY CARE 
M. Alfred Haynes, M.D., Girma Wolde-Tsadik, Ph.D. 
and Paul Juarez, Ph.D. 



INTRODUCTION 

A corrparison of the health status of various ethnic groups must be 
approached from several different perspectives because no single data 
source provides a fully satisfactory picture. Since birth and death 
events are more widely and reliably recorded, vital statistics are 
frequently quoted. The differences in mortality are easily summarized as 
differences in life expectancy, and one can easily grasp the significance 
of wide gaps in longevity. For example, in 1983, life expectancy at 
birth for the White population was 75.2 years while for the Black 
population, it was 69.6 years. ^ But even this basic information is not 
always available for all ethnic groups. Trevino has noted that "at 
present we do not even know how many Hispanics die each year in this 
country, let alone their health status, use of services or unmet health 
care needs. "^ 

Another approach to corrparing health status would be to use institutional 
data such as that of the National Hospital Discharge Survey-^ or the 
National Nursing Home Survey.'^ This approach reflects only the 
experience of institutionalized populations and, therefore, has certain 
limitations. A much broader picture could be obtained from the National 
Health Interview Survey which is a nationwide survey conducted through 
personal household interviews. Data are collected on illnesses, 
injuries, impairments, chronic conditions, the use of health resources 
and a variety of other health and health-related characteristics.^ 
These data are subject to many limitations including problems about the 
accuracy of diagnosis and the fact that respondents cannot address 
problems of which they are not aware. On the other hand, the National 
Health and Nutrition Examination Survey^ obtains data by means of 
direct physical examinations along with clinical and laboratory tests. 
The sairple size is much smaller than that of the National Health 
Interview Survey. 

An alternative method is the collection of data from a sample of 
physicians in office-based practices. This is the approach of the 
National Ambulatory Medical Care Survey, "7 a continuing national 
probability sairple of ambulatory medical encounters. The survey does not 
include visits to hospital-based physicians or to specialists in 
anesthesiology, pathology or radiology. Non-office visits and telephone 
contacts are not incorporated. 



107 



In order to obtain a true picture of the differences of health status of 
various groups, it is necessary to examine indicators from all of these 
surveys and to draw conclusions on the basis of the conposite picture. 
Each analysis adds another dimension of significance. The purpose of 
this study was to examine the differences from the perspective of 
ambulatory care and to complement other analyses under consideration by 
the Task Force on Black and Minority Health. Inasmuch as most of the 
contacts with the health care system are made in the context of 
ambulatory care, this study should be considered as one of the irrportant 
elements of the total picture. It differs from previous studies in 
ambulatory care which focus on other aspects. For example, there are 
studies of the patterns of ambulatory care visits to surgeons, ^ 
obstetricians^ and family practitioners. ^^ Also, there have been 
studies on specific groups such as the Asian Pacific Islanders^l or, in 
general, the poor.-'-^ This study, on the other hand, is based on a 
conparison of the major racial/ethnic groups — White, Black, Hispanic, 
Asian and Native American. 

METHODS 

The data source was the 1981 National Ambulatory Medical Care Survey that 
included 43,123 patient visits. The frequency of patient encounters by 
the various racial/ethnic, sex and ten roost frequent diagnostic groups 
was analyzed using the multiway contingency table method based on the log 
linear model. The basic chi-square technique for two-factor analysis was 
also employed in the preliminary examination of association between 
race/ethnicity and disease conditions. In addition, the importance of 
certain disease-ethnicity associations was highlighted by calculating the 
percentage contribution of these specific associations to the overall 
significance of the findings. 

The small number of visits of minority patients did not allow the 
performance of epidemiological ly meaningful analyses by age groups. In 
fact, the 66 patient encounters by Native Americans were too few to 
incorporate in any one of the analyses. However, the frequency 
distribution of Native Americans by diagnostic categories will be 
presented. In order to maintain the minimum allowable expected cell 
frequencies. Black and White Hispanics were grouped to form a single 
category. 

FINDINGS 

The preliminary description of the data consists of a cross-tabulation of 
18 diagnostic groups with the five racial/ethnic categories. The 
frequency counts are given in Table 1. More than 20% of the cells do not 
satisfy the minimum expected cell frequency of 5. The collapse of the 
relevant diagnostic categories, the exclusion of the Native American 
group and the subsequent performance of a test of association reveals a 
highly significant relationship between the health problems and the 
racial/ethnic groups (p^.OOOl). 



108 



In Table 2, the results of the test of association are presented in a 
reduced form. The diagnostic categories constitute those conditions that 
show more than equal contribution to the overall association. The 
entries are the percentage of the total chi-square attributable to the 
given cell. The plus and minus signs, respectively, indicate the 
observed frequencies that are more and less than the expected values 
under independence of the two factors. All ethnic groups showed 
significant association with some diagnostic category. The VJhite 
patients were more likely to present themselves at ambulatory care 
settings with psychiatric problems than the other groups. Blacks and 
Hispanics, in that order, were far less likely to seek care for these 
conditions. Asian Americans were more frequently seen for diseases of 
the nervous and sensory systems, whereas Blacks and Hispanics were less 
likely to present themselves with these problems than would be expected. 
Among Blacks, a high frequency of problems of the circulatory system 
contributes to the overall significance of the relationship between 
diagnosis and race/ethnicity. Hispanics show a substantially lower 
frequency of visits for circulatory problems and, to a lesser extent, 
Asians account for infrequent care of this category. Respiratory 
problems are frequent causes of visit for Blacks as are digestive 
diseases for Hispanics. Musculoskeletal diseases, as a cause of visit 
for Hispanics, are major contributors to the disease and ethnicity 
relationship v;hereas Asians are less likely than expected to seek care 
for this set of conditions. Visits for the diagnostic category 
classified as "supplementary" are strongly associated with Asians. The 
classification is, generally, described as factors influencing health 
status and contact with health services. More specifically, it 
enumerates health hazards related to communicable diseases and family 
history as well as familial conditions related to social, mental and 
economic well-being. 

Finally, in Table 3, we present a key result of the three-way contingency 
table analysis. The three factors constitute sex, race/ethnicity and the 
ten most frequent diagnostic categories. The latter incorporate the 
conditions identified in Table 2 as well as injury/poisoning, 
genitourinary and skin diseases. The table entries represent the partial 
associations due to the indicated factors after the effect of the missing 
factor is removed. Hence, the rows of primary interest are the fourth, 
fifth and sixth. As demonstrated by the p-values, all the paired partial 
associations are significant. The findings presented in Tables 1 and 2 
are further substantiated by the significance of the ethnicity-diagnosis 
partial association. It should be noted, however, that those results 
were based on more diagnostic categories and the marginal distributions 
of the frequency of visits. A more detailed scrutiny of the 
sex-diagnosis association reveals a more or less general pattern across 
ethnic groups. This pattern encompasses a disproportionately high number 
of male visits for genitourinary health problems and those of the 
supplementary classification, and an unexpectedly high number of female 
visits for injury and poisoning. The common configuration appears in 
Wiites, Blacks and Hispanics. The latter also reveal a relatively high 
frequency of female visits for diseases of the digestive system. The 



109 



Asian group did not show any strong association between health problems 
and sex. The remaining results of Table 3 are of academic interest. The 
first three rows substantiate the disproportionate number of male, White 
and certain disease categories in the survey sample. The last row of 
sex-ethnicity-diagnosis with the non-significant result (p=0.1852) merely 
confirms the appropriateness of the log linear model. 

DISCUSSIONS 

The interpretation of the results should be characterized by caution. 
People with different attitudes towards health present themselves for 
care at different levels of severity. The high frequency of visits by 
Whites for psychiatric conditions, and those by Asians for diseases of 
the supplementary classification might reflect the groups' greater access 
to care and/or a greater sense of medical need. 

The supplementary classification includes preventive services but the 
significance of this finding cannot be established due to the relatively 
small number of patient visits. On the other hand, if Blacks and 
Hispanics feel a lesser sense of medical need, then the magnitudes of the 
identified circulatory, digestive and musculoskeletal problems are 
underestimated and there might be other problems to which they are at 
greater risk but \/hose significance is suppressed. This is of special 
interest since the morbidity findings are certainly not as striking as 
the mortality statistics would suggest. 

Some of the findings suggest the need for more detailed studies with 
respect to age and sex. For exartple, the disproportionately high number 
of female visits for injury and poisoning in Whites, Blacks and Hispanics 
is not exactly what one would have expected. The absence of association 
between health problems and sex in Asians also deserves attention due to 
its possible inplications for the other groups. 

In summary, this study, despite its limitations, has established strong 
associations among the factors of sex, race/ethnicity and health problems 
as reflected in ambulatory care settings. A more detailed study of a 
larger saitple of minority populations would provide additional 
information on these findings by permitting further analysis of more 
specific age categories. 



110 



TABLE 1: FREQUENCY OF DIAGNOSIS BY RACE AND ETHNICITY 









Race/Ethnicity 






Principal Diagnosis 


Non-Hispanic 


Hispanic 


Asian 


Native 
American 


Total 




White 


Black 




Infective & Parasitic 


1,006 


123 


51 


13 


7 


1,200 


Malignant Neoplasms 


685 


49 


20 


6 





760 


Other Neoplasms 


385 


37 


15 


2 





439 


Endocr./Nutri./Metabol. 


1,204 


155 


73 


9 





1,441 


Blood/Blood Forming Organs 


199 


26 


8 


5 





238 


Psychoses & Neuroses 


2,816 


148 


86 


21 


1 


3,072 


Nervous & Sensory Systems 


3,419 


273 


132 


51 


5 


3,880 


Circulatory 


3,679 


452 


123 


17 


4 


4,275 


Respiratory 


4,037 


495 


230 


41 


11 


4,814 


Digestive 


1,614 


185 


110 


17 


3 


1,929 


Genitourinary 


2,351 


248 


137 


16 


3 


2,755 


Diseases of the Skin 


1,711 


145 


75 


16 


2 


1,949 


Musculoskeletal 


2,704 


271 


212 


15 


4 


3,206 


Congenital Anomalies 


196 


17 


13 


4 





230 


Symptoms 


1,247 


147 


81 


8 





1,483 


Injury & Poisoning 


3,016 


343 


166 


29 


5 


3,559 


Supplem Class 


6,433 


610 


319 


90 


21 


7,473 


Other Causes 


142 


30 


18 


7 





197 


Unrecorded 


188 


25 


8 


2 





223 



TOTAL 



37,032 3,779 1,877 



369 



66 



43,123 



111 



TABLE 2: STRENGTH OF ASSOCIATION BETWEEN DIAGNOSES AND ETHNICITY 

(CELL PERCENT CONTRIBUTION*) 





Race/Ethnicity 


Principal Diagnosis 


Non- 


-Hispanic 


Hispanic 






White 


1 Black 


Asian 


Psychoses and Neuroses 


2.8 
( + ) 


1 12.8 
1 (-) 


4.0 
(-) 




Nervous and Sensory Systems 




1 3.1 
1 (-) 


1.9 
(-) 


2.2 
( + ) 


Circulatory 




1 3.7 

1 { + ) 


5.0 
(-) 


2.5 
(-) 


Respiratory 




1 3.0 
1 ( + ) 






Digestive 






1.9 
(+) 




Musculoskeletal 






8.8 
( + ) 


1.3 
(-) 


Suppleirientary Class 








2.5 
( + ) 



* -A plus and minus sign, respectively, indicate more and less 
expected frequencies under independence. 
-Enpty cells have values less than or equal to 1.25 percent. 



than 



112 



TABLE 3: PARTIAL ASSOCIATION OF FACTORS 





Degree 








of 
Freedom 


Partial 


Association 


Factors 


Chi-Square 


P-Value 


Sex 


1 


1558.7 


0.0000 


Ethnicity 


3 


67576.3 


0.0000 


Diagnosis 


10 


6456.2 


0.0000 


Sex-Ethnicity 


3 


30.1 


0.0000 


Sex-Diagnosis 


10 


1088.0 


0.0000 


Ethnicity-Diagnosis 


30 


277.2 


0.0000 


Sex-Ethnicity-Diagnosis 


30 


36.7 


0.1852 



113 



REFERENCES 



1. National Center for Health Statistics: Health, United States, 
1984 . DHHS Pub. No. (PHS) 85-1232, Public Health Service, 
Washington, U.S. Government Printing Office, Dec. 1984. 

2. Trevino, P.M.: Vital and Health Statistics for the U.S. Hispanic 
Population. American Journal of Public Health , Sept. 1982; 7(9): 
979-82. 

3. National Center for Health Statistics: Utilization of short-stay 
hospitals, annual summary for the United States, 1982, by E.J. 
Graves, Vital and Health Statistics , Series 13-No. 78, Public Health 
Service, Washington, U.S. Government Printing Office. 

4. National Center for Health Statistics: The National Nursing Home 
Survey, 1977 summary for the United States, by J.F. Van Nostrand, A. 
Zappolo, E. Hing, et. al.. Vital and Health Statistics , Series 
13-No. 43, DHHS Pub. No. (PHS) 79-1794, Public Health Service, 
Washington, U.S. Government Printing Office, July 1979. 

5. National Center for Health Statistics: Current estimates from the 
National Health Interview Survey, United States, 1981, by B. Bloom, 
Vital and Health Statistics , Series 10-No. 141, DHHS Pub. No. (PHS) 
82-1569, Public Health Service, Washington, U.S. Government Printing 
Office, Oct. 1982. 

6. National Center for Health Statistics: Plan and Operation of the 
second National Health and Nutrition Examination Survey, 1976-80, by 
A. McDowell, A. Engel, J.T. Massey, and K. Maurer, Vital and Health 
Statistics , Series 1-No. 15, DHHS Pub. No. (PHS) 81-1317, Public 
Health Service, Washington, U.S. Government Printing Office, July 
1981. 

7. National Center for Health Statistics: 1981 Summary, National 
Ambulatory Medical Care Survey, by L. Lawrence and T. McLemore, 
Advance Data for Vital and Health Statistics , No. 88, DHHS Pub. No. 
(PHS) 83-1250, Public Health Service, Hyattsville, Md., Mar. 16, 
1983. 

8. cypress, B.K.: Patterns of ambulatory care in office visits to 
general surgeons: The National Ambulatory Medical Care Survey, 
United States, January 1980 - December 1981. Vital Health 
Statistics [13] , Sept. 1984; 79:i-iv, 1-61. 

9. Cypress, B.K. : Patterns of ambulatory care in obstetrics and 
gynecology: The National Ambulatory Medical Care Survey, January 
1980 - December 1981. Vital Health Statistics [13] , Feb. 1984; 
76:1-62. 



114 



10. Cypress, B.K. : Patterns of ambulatory care in general and family 
practice: The National Ambulatory Medical Care Survey, United 
States, January 1980 - December 1981. Vital Health Statistics [13] , 
Oct. 1983; 75:1-60. 

11. Yu, E.S. and Cypress, B.K.: Visits to physicians by Asian/Pacific 
Americans. Medical Care , Aug. 1982; 20(8) :809-20. 

12. Kleinman, J.C, Gold, M. and Makuc, D.: Use of Ambulatory Medical 
Care by the Poor: Another Look at Equality. Medical Care , Oct. 
1981; 19(10): 1011-29. 



115 



Nutritional Status and 
Dietary Pattern of Racial 
Minorities in the 
United States 



Shiriki Kumanyika, Ph.D., R.D. 

Assistant Professor 
Department of Epidemiology 
The Johns Hopkins University 
School of Hygiene and Public Health 
Baltimore, Maryland 

Deborah L. Helitzer 

Doctoral Candidate 
Department of International Health 
The Johns Hopkins University 
School of Hygiene and Public Health 
Baltimore, Maryland 



ABSTRACT 



Dietary patterns of minority groups may differ from those of 
the general population according to several factors. These factors 
include the nature of the original diet, the ways the diet has been 
adapted to or supplanted by dietary patterns from the dominant U.S. 
culture, availability of preferred foods, and acculturation. The 
objectives of this review have been to describe general themes and 
arphases in the diets of Asian, Black, Hispanic and Native Americans 
and, where possible, to identify associations between these dietary 
patterns and excess nutrition-related health risks. The content of 
the review has been affected by the uneven availability of data for 
different subgroups and by certain coirplexities which are associated 
with cross-cultural interpretation of dietary and nutritional status 
data. 

Following a general discussion of relevant nutrition concerns 
and data issues, specific dietary and nutritional concerns for each 
minority subgroup are sunmarized. In general, nutritional factors 
which may contribute to health disparities between minorities and 
whites can be grouped into four categories, as follows: 1) excess 
risks related to the low-income status of a high proportion of 
minority individuals and families; 2) risks related to the excess 
prevalence of obesity in some minority groups; 3) carryovers of 
dietary risks from traditional diets; and 4) increased risks related 
to movement away from traditional diets that are low in fat and 
cholesterol and high in conplex carbohydrates to an opposite, more 
typically American, pattern. 



118 



CONTENTS 
LIST OF TABLES Al© FIGURES 

1. INTRODUCTION 

2. BACKGROUND 

2.1 Sources of Data on the Diets and Nutritional 

Status of Minority Groups 

2.2 Nutrition-Related Health Variables 

2.3 Methodological Issues 

3. ASIAN AMERICANS 

3.1 General dietary patterns 

3.2 Nutritional risk 

3.3 Conclusions 

4. NATIVE AMERICANS 

4.1 General dietary patterns 

4.2 Nutritional risk 

4.3 Conclusions 

5. HISPANIC AMERICANS 

5.1 General dietary patterns 

5.2 Nutritional risk 

5.3 Conclusions 

6. BLACK AMERICANS 

6.1 General dietary patterns 

6.2 Nutritional risk 

6.3 Conclusions 

7. SUMMARY 

LITERATURE CITED 

SUPPLEMENTARY REFERENCES 

APPENDICES 

i^pendix 1: Background Information on Vitamins and Minerals 
Appendix 2: Recommended Dietary Allowances and Estimated Safe 

and Adequate Daily Dietary Intakes 



119 



LIST OF TABLES AND FIGURES 



Table 


1: 


Table 


2: 


Table 


3: 


Table 


4: 


Table 


5: 


Table 


6: 


Table 


7: 


Table 


8: 


Table 


9: 


Table 


10: 


Table 


11: 


Tabel 


12: 


Figure 1: 



Current Nutrition Concerns for U.S. Children 

Current Nutrition Concerns for U.S. Adults 

A Summary of Dietary Determinants Related to the 
Nutritional Risk Status of U.S. Minority Groups 

Characteristic Chinese Foods and Food Choices 

Characteristic Japanese Foods and Food Choices 

Characteristic Filipino Foods and Food Choices 

Exairples of Some Southeast Asian Foods 

Examples of Traditional and Contemporary Foods 
in the Diets of American Indians 

Characteristic Mexican-American Foods and Food 
Choices 

Characteristic Puerto Rican Foods and Food 
Choices 

Characteric Black Foods and Food Choices 

Exanples of Some Caribbean Foods 

Percent of U.S. adults ages 20-74 who are 
overweight, by race and sex 



120 



1. INnOXXJTION 

Minority groups in the l&iited States coirprise several subgroups 
whose traditional dietary patterns may differ from those of the 
general population. The diets of minority group persons are 
influenced by several, interrelated factors, for exanple, the nature 
of the traditional diet, ways in which the diet has been adapted to 
or supplanted by dietary patterns from the dominant U.S. culture, the 
availability of preferred foods, and acculturation. The contribution 
of culturally-determined dietary patterns to disparities in morbidity 
and mortality relates both to protective features of the original 
diets which may have been discontinued as well as to potentially 
harmful practices which may persist or have been acquired during 
acculturation. 

Minority-specific factors operate within the larger context of 
universal determinants of dietary intake and nutritional need. These 
universal determinants, which apply in all societies, include factors 
which are biological (e.g. heredity, gestation, growth, energy 
expenditure, illness) , psychological (mental state, knowledge, 
attitudes) , socio-cultural (religion, household structure, child 
rearing practices, family interactions, food beliefs) , and 
socioeconomic (per capita food availablity, purchasing power, 
dependency, stability) in nature. 

The Secretary's Task Force on Black and Minority Health was 
convened specifically to identify aspects of the environment which 
predispose minority groups to disparities in health and to reconmend 
relevant policy. Therefore, the Task Force deliberations provide a 
very appropriate context for reviewing diet and nutrition literature 
related to minority groups. 

The objective of this review has been to describe, in very 
general terms, the themes and emphases in the diets of various U.S. 
minority sub-groups and, where possible, to identify apparent 
associations between these dietary patterns and excess 
nutrition-related health risks. The specific questions addressed 
here are whether Black, Hispanic, Asian, or Native Americans are at 
excess nutritional risk and whether this excess risk can be 
attributed to culturally- rather than to socioeconomically-based 
dietary factors. Coirplex issues are involved. Technical aspects are 
acconpanied by extremely sensitive social and economic concerns. 
Some exanples follow: 

1) Pood patterns are an integral aspect of culture. Retention 
of a traditional food pattern is an expression of ethnic 
identity. Thus, evaluation of minority group diets irust be 
dDjective and must be sensitive to perceptions that the 
underlying culture is being criticized or patronized; 



121 



2) Although diets of mentoers of a particular minority group 
may be more similar to each other than to those of the 
general population r there is a great deal of 
heterogeneity. Insufficient attention to intracultural 
diversity will lead to incorrect conclusions and may be 
viewed as stereotyping; 

3) Findings of undernutrition among the low-income segment of 
the minority population may be attributed to inadequacies 
in the food and nutrition programs designed to attentuate 
poverty-associated malnutrition. In addition, such 
findings may be attributed to incompetence of 
low-income minority individuals and families. From a 
social policy point of view, both of these attributions are 
quite sensitive. As recently pointed out by the President's 
Task Force on Food Assistance (1983) , documentation of even 
a modicum of undernutrition amidst the abundance of food in 
the U.S. is a sign of societal failure at some level, 

4) Findings of obesity and excess risk of sane nutrition- 
related chronic diseases among minorities suggest that 
adaptation to U.S. dietary patterns may have 
disproportionately deleterious consequences for 
minority group persons; 

In the text which follows, we have noted the nutrition concerns 
which we consider relevant to current issues of minority health, have 
listed some inference issues related to assessment of nutrition in 
non-white populations, and have summarized pertinent literature in 
this context. We have been especially aware of the tremendous gaps in 
the nutrition literature in this respect and of the outdatedness of 
much of the literature which is available. We have exphasized 
studies dated 1975 or later. However, even studies published within 
the last decade may not accurately represent those aspects of 
nutritional status which respond to changes in levels of food and 
nutrition program funding and outreach. 



122 



2. BACKGROUND 

2.1 Sources of Data on the Diets and Nutritional Status 

of Minority Groups 

The availability of timely, representative dietary and 
nutritional status data varies greatly for the different U.S. 
minority sub-populations in rough proportion to the sub-group 
population size. National probability sairple estimates for 
nutrition-relevant variables are available for blacks from the 
National Health and Nutrition Examination Surveys, 1971-75 and 
1976-80 (NHANES I and NHANES II) and from the National Pood 
Consuirption Survey 1977-78 (NFCS) and will be available for Mexican, 
Puerto Rican, and Cuban Americans from the Hispanic Health and 
Nutrition Examination Survey (HHANES) coirpleted in 1985. Special 
surveys of the NFCS were also conducted in Alaska, Hawaii, and Puerto 
Rico, permitting — at least theoretically — estimates for minority 
sub-populations in these areas. 

NHANES data provide four types of nutritional status measures 
for each sairple person (anthropometric, clinical, biochemical, and 
dietary) as well as accoirpanying health indicators (medical history, 
dental and physical health by examination) . The NFCS (and some other 
subdivisions of the Consumer Food and Economics Institute) provides 
detailed data on individual and household food consunption patterns 
according to household characteristics. At present, one of the major 
shortcomings of these surveys results from the substantial time lag 
between data collection and publication. While these data sources 
have been useful for monitoring over relatively large time intervals, 
this time lag renders them unsuitable for assessment of acute 
nutritional status, short-term, or very recent changes. 

The Ten-State Survey, often cited in the nutrition literature, 
was conducted in 1968-69 and is now conpletely out of date. Ten 
State findings should be used only as a baseline for the nutritional 
status of vulnerable groups prior to inplementation of the expanded 
base of food and nutrition programs (program expansions were in large 
part a response to Ten-State Survey findings) . 

With the exception of HHANES, the ability to assess 
race-specific nutritional vulnerability is not assured by the NHANES 
sarrpling design. The sampling ctojective is to permit representative 
estimates for the overall U.S. population rather than to create a 
data base for sub-group comparisons. Although sub-groups at risk due 
to low income, reproductive status (women of childbearing age) , and 
age (preschool children and the elderly) have been oversairpled in 
these surveys, sairple sizes for racial subgroups do not necessarily 
meet criteria for stable estimates — especially if cross-classified by 
sex, age, and socioceonomic indicators. 



123 



Separate tabulations for blacks are frequently but not routinely 
reported in NEIANES publications. Accoirpanying statistical analyses of 
the significance of black-white differences are sometimes included. 
Data for minority groups other than blacks are either excluded from 
sub-group tabulations or included with data for blacks under a 
general heading of "non-^hite". Additional information on blacks 
can often be gleaned from publisl^d reports of special analyses by 
National Center for Health Statistics staff or from published 
analyses of NHANES data by manbers of the research ccamiunity at 
large. Fewer relevant reports based on MFCS data than on NHANES data 
were identified during the preparation of this background paper. The 
apparent amount of NFCS data is larger than the actual data base 
which can be used for racial coirparisons, because tables with racial 
breakdowns are often not age- and sex-adjusted. 

Another major source of population data on nutritional status is 
the Centers for Disease Control Pediatric and Pregnancy Nutrition 
Surveillance System (CDC-PNSS) (Centers for Disease Control, 1983). 
This system coitpiles height, weight, and iron status data for 
children and women in publicly-supported health programs in more than 
20 states. CDC reports include separate tabulations for black, 
Hispanic, Native American and Asian children (including Southeast 
Asian refugees) , using measuremoits taken upon entry into a nutrition 
or health program. These data are suitable for racial conparisons 
within the data base, but they cannot support gaieralities about 
minority children in the general population, or even to low- income 
minority group children. Participation in the CDC-HiSS is voluntary 
and varies by state, county, and program. 

Additionally, minority groups are variably represented in 
prevalence data derived from baseline screenings for cardiovascular 
disease studies (e.g.,Framingham, Lipid Research Clinics, Honolulu, 
Puerto Rico, and San Antonio Heart Studies) , in data from evaluations 
of nutrition programs (such as School Ijanch, the Supplonental Feeding 
Program for Women, Infants, and Children, Head Start, or the 
Nutrition Program for Older Americans) and in small studies on 
various nutritional status issues. Unfortunately, for the numerically 
smaller minority groups, the non-coirprehensive and non-representative 
data sources are often the only basis for describing possible 
nutritional disparities. This unevenness of coverage is clearly 
evident in the later sections of this paper. 

2.2 Nutrition-Related Health Variables 

The undernutrition- and chronic disease-related nutritional 
concerns addressed in this paper are suinnarized in Tables 1 and 2. 
For reference, listings of essential vitamins and minerals and 
recommended nutrient intakes have been appended (see Appendices 1 and 
2) . Tables 1 and 2 are not comprehensive listings of nutrition and 
health concerns. Rather, the prd^lems indicated are those most 
relevant to current public health policy. 



124 



Table 1: 

CXirrent Nutrition Concerns for U.S. Children 



LIFE CYCLE 
STAGE 

GEsnaiou, 

INFANCY, 
CHILDHOCO, and 
ADOLESCENCE 



DIET and NUTRITION 
VARIABLES 

- inadequate intakes 
of energy and other 
essential nutrients 



overfeeding, feeding 
of highr<:alorie-low 
nutrient-density 
foods 



ASSOCIATED HEALTH 
OOTCOHES 

- low birth weight, 
developniaital or 
functional inpairment, 
growth stunting, iron 
deficiency aneonia, low 
resistance to infection, 
poor general health 

obesity, poor food habits, 
predisposition to obesity 
and obesity-related 
diseases 



between-meal snacks, 
sugary snacks 



- dental caries 



SOURCES: Food and Nutrition Board, 1980; National Institutes of 
Health, 1982 



Conceptual and methodological limitations preclude definitive 
answers to the types of nutrition and health questions that are 
currently relevant to U.S. population groups — minority group issues 
included. At present, the occurrence of severe nutritional 
deficiencies in U.S. populations is limited to certain unusual 
circumstances. The most pertinait undernutrition issues relate to 
subtle effects of subclinical nutrient deficiencies on health outcomes 
at various stages of the life cycle. Most of these outcomes are not 
nutrition-specific, i.e. they are influenced by other biological and 
environmental factors. The pertinent ovemutrition issues are defined 
in terms of consistent, although not firmly established, associations 
of various dietary excesses with chronic disease risk factors or 
outcomes — again not nutrition-specific outcomes. 



125 



Table 2: 

Current Nutrition Concerns for U.S. Adults 



LIFE CYCLE 
STAGE 

REPRODUCTIVE 
PERIOD 
(Girls fWbmen) 



DIET and NUTRITION 
VARIABLES 

inadequate intakes of 
energy and other 
essential nutrients 



ASSOCIATED HEALTH 
OUTCOMES 

poor preconceptual 
nutrition, poor pregnancy 
outcomes, poor lactation 



ADULTHOOD 



- inadequate intakes of 

energy and other 
essential nutrients, 
drug-induced nutrient 
deficiencies 

- excess calories 



poor general health and 
functional status, 
nutritional anenias 



- excess sodium 

- excess smoked or 

pickled foods 

excess saturated fat 
and cholesterol 

excess saturated fat 

suboptimal vitamin A* 

suboptimal vitamin C 

suboptimal potassium 

suboptimal fiber 

suboptimal calcium 



- obesity and dDesity-related 

risk of diabetes, heart 
disease, endometrial and 
gallbladder cancers 

- increased hypertension risk 

- increased hypertension and 

gastric cancer risk 

- increased heart disease risk 



increased breast cancer risk 

increased lung cancer risk 

increased esophageal cancer 
risk 
increased hypertension risk 

increased colon cancer risk 

increased osteoporosis risk 



* high vitamin intakes have been been associated with prostate cancer 
(Graham et al., 1983; Biterline, 1984); 

SOURCES: Food and Nutrition Board, 1980; American Heart 
Association, 1982; Dietary Guidelines for Americans, 1980; 
Committee on Diet, Nutrition, and Cancer, 1982; and Willet and 
MacMahon, 1984; National Institutes of Health, 1984 



126 



In low-income groups, problems of undernutrition and chronic 
disease risk may coexist. For exairple, Trovtoridge has pointed out 
the excess prevalence of both linear growth stunting and obesity 
among Native American and Hispanic children in the 1982 Centers for 
Disease Control Surveillance System. The data indicate that the diets 
of these groups may be quantitatively adequate but qualitatively 
deficient in some respects. High quality protein or essential 
vitamins and minerals may be lacking in the diets of these children 
relative to the level of calories available (Trowbridge, 1983) . An 
additional exanple of this possible coexistence of dietary inadequacy 
and dietary excess relates to sodium intake. The addition of salt or 
other high sodium seasonings (soy sauce, canned chicken broth, 
bouillon, salt pork) enhance the flavor but not the nutritional value 
of foods. A dietary pattern may be both nutritionally inadequate and 
excessive in sodium content. 

2.3 Methodological Issues 

Many diet and nutritional status assessments rely on self -report 
data or physical measurements which vary widely within each 
individual from day to day as well as between individuals. 
Representative measurements of groups may not be strictly repre- 
sentative of any individual within the group. An additional factor 
which limits the ability to generalize about nutriticmal status is 
the lack of a single dietary or nutrition variable that can be used 
to indicate risk across all dietary or nutritional dimensions. 

No one dietary pattern is inherently superior to another. The 
flexibility, elasticity, and feedback controls inherent in biologic 
systems permit the attainment of nutritional status equivalence 
through intakes of a wide variety and combination of foods — as 
demonstrated by the diversity of nutritionally adequate diets across 
cultures throughout the world. Nutrition problems idaitified on the 
basis of food guides (e.g. the U.S. basic four food groups) or 
recomnended dietary allowances should be interpreted cautiously. 
Dietary adequacy is best evaluated by a combination of dietary 
measures and direct physiologic indicators of nutriticaial status. 
The appropriateness of applying certain dietary and physiological 
reference standards which are based on the U.S. population at large 
to minority populations is also an issue. 

Some exairples of minority-specific nutrition methodology issues 
follow. 

a. The precision of dietary assessment methods decreases 
with increasing dietary diversity or instability; this may 
pose a particular problem of dietary assessment among 
minority group members whose food habits are evolving in 
response to migration or other social and environmental 
factors. 



127 



b. Conventional methods of dietary assessment and 
evaluation in the United States are based on the 
contiinations of foods and food preparation practices connon 
to the majority. These methods may not reflect iirportant 
dimensions on which diets of minority groups differ from 
more typical U.S. diets (Jerome and Pelto, 1981; Sanjur, 
1982) . Calculated intakes of nutrients do not reflect the 
foods and conbinations of foods from which the nutrients 
were derived, nor do they reflect certain features of food 
preparation which can affect biological availability of 
nutrients or actual nutrient content. 

For exanple, the presence of meat or vitamin C in a 
meal increases overall iron absorption from that meal (Food 
and Nutrition Board 1980) ; tea and certain elements in foods 
may reduce absorption of iron or other minerals; soaking 
bcHies in acid solution (a step in some Chinese soup 
preparation) increases the calcium content of the soup 
(Suitor and Crowley, 1982) ; consuiiption of foods prepared 
from lime-treated corn adds significantly to dietary calcium 
and to niacin availability in some Mexican and American 
Indian cultures (Roe, 1973) ; use of iron cooking utensils 
adds to dietary iron; the practice of chewing bones in many 
cultures provides nutrients which are not reflected in 
calculations based on fleshy parts of the meat, 

c. There is substantial evidence that the distributions of 
hemoglobin are different for blacks and whites (Owen and 
Yanochik-Owen,1977; Meyers et al., 1979; Koh et al. 1980b, 
Frerichs et al. 1977) — the black distribution is 
approximately one mg/dl lower — ^parently related to the 
relationship between hemoglobin and transferrin saturation 
(Meyers et al . , 1979) . This difference is independent of 
iron intake. The prevalence of iron deficiency anemia among 
blacks may be overstated if the "normal" range for 
hemoglobin levels is calculated from white population data 
(Owen and Yanochik-Oven, 1977) . 

d. The applicability of reference data for skeletal and 
dental development is influenced by racial differences in 
patterns of skeletal and dental development. As sumnarized 
by Knishbacher (1983) black infants tend to be smaller, but 
developmentally advanced at birth; black boys and girls are 
taller than their white counterparts during the 2 to 14 year 
age range, bone density is greater and tooth eruption 
earlier among black vs. white children (Gam and Clark, 
1976; Kramer et al., 1977; Shutte, 1980). Use of white 
population standards may lead to overestimates of birth 
weight-associated nutrition risks and underestimates of 
linear growth stunting. 



128 



e. The level of health risk at a given level of exposure 

to nutrient excess nay vary on a genetic basis. For 

exairple: blacks may be more sensitive than whites to a given 
level of sodium exposure due to evolutionary factors 

favoring renal sodium conservation (Gillum, 1979) ; a 
"thrifty gene" favoring the survival of individuals with 
maximum fat stores during times of food scarcity has been 
proposed as a factor in the excess prevalence of obesity and 
diabetes among Pima Indians (Sievers and Fisher, 
1981) . 

f . Some environmental factors and health problems which 
increase nutritional risk are in excess prevalence among 
minority groups, e.g., a greater proportion of black than 
white men are cigarette smokers (Naticaial Goiter for Health 
Statistics, 1983a; Neaton et al., 1984); alcohol use is more 
prevalent among some minority sub-groups vs. the goieral 
population (Sievers and Fister, 1981) Strimbu and Sims, 
1974; the major chronic diseases which are aggravated 

by dietary excesses are in excess prevalence among minority 
groups, e.g., hypertension and diabetes among black 
Americans; diabetes among Mexican Americans and American 
Indians (refer to other sections of the Task Force report) . 
Thus, the requirements for "normal" nutrition of minorities 
may be misspecif ied by standards designed for populations 
with different patterns of health-related nutritional 
risk. 

g. Dairy product use, particularly milk drinking, is 
traditionally lower in non-white populations than white 
populations after infancy — a pattern thought to be related 
to the higher prevalence of lactose intolerance in non-white 
vs. white children and adults after infancy (Paige et al., 
1971) . However, assuirptions about calcium intake based on 
presumed lactose intolerance should be avoided. Patterns of 
milk drinking among non-white minority groups indicate that 
lactose intolerance is not equivalent to milk intolerance; 
studies among lactose intolerant vs. tolerant children and 
adults have not always <±)served significant differences in 
milk consunption — ^particularly when the milk is offered in 
conjunction with subsidized feeding programs programs 
(Stephenson et al.,1977 Newcomer et al., 1977, Marrs, 

1978) . 

In addition, evaluation of the adequacy of calcium 
intake solely on the basis of dairy product intake should be 
avoided. Dairy products are the best, but not the only 
sources of calcium. Traditicaial calcium sources among 
non-white populations include tofu (soybean curd) and small 
bony fish in Asian diets, leafy green vegetables in the 
diets of southern Blacks, and beans and lime-treated com 
in Hispanic diets. The density and bioavailablity of 
calcium in these foods varies; however, since calcium 



129 



absorption is higher at lower levels of calcium intake and 
varies with diet conposition and other factors (Avioli, 
1980) , calcium intake may not be a good indicator of calcium 
status except where dietary deficiencies are extreme. The 
use of number of "milk group" servings according to the U.S. 
"basic four" food guide is misleading as a standard for 
evaluating calcium intake in individuals from traditionally 
non-milk-drinking cultures. 

Further, adequacy of a given calcium intake may be 
different for different ethnic groups. There has been 
substantial interest in the possible role of chronic low 
calcium intakes in the incidence of osteoporosis among 
postmenopausal U.S. women (National Institutes of Health, 
1984) . However, even if the calcium-osteoporosis 
association exists in white women, there appear to be other, 
yet undetermined factors which afford relative protection 
for some non-white women against osteoporosis. The 
associated relative risks are substantially higher for vAiite 
women than for black women, (Farmer et al., 1984) (although 
not necessarily higher than for all non-white women (Yano et 
al., 1984)). This is in spite of calcium intakes among 
black women which are lower than those of white women and 
which are often below the accepted U.S. standard of dietary 
calcium adequacy. 

h. Dietary intake is influenced by socieconomic status. 
Nutrient intakes are higher at higher levels of disposable 
income (in the low to middle income range) , with the 
exception of carbohydrate intake, which decreases with 
increasing income (Adrian and Daniel, 1976) . This 
complicates the separation of culturally- and 
socioeconomically-based nutritional differences in data 
which are not income-adjusted. Similarly, income-related 
differences in access to health care coirplicate the 
interpretation of nutritional status measures which are 
sensitive to medical intervention (e.g. prevalence of low 
birth weight, dental care) . 

In concluding this background section, it is useful to give a 
general overview of the factors which determine dietary quality for 
minority groups within the U.S. environment. This overview (table 3) 
avoids the redundancy of specifying these factors separately for each 
minority group, and at the same time draws attention to coinnon 
pathways of nutritional risk. Also, considering the limited basis 
for making quantitative judgenents on issues of interest, this 
overview is intended to facilitate qualitative inferences about 
potential areas of concern. 



130 



Table 3: 

A Suirmary of Dietary Determinants Related to the Nutritional Risk 

Status of U.S. Minority Groups 



> culture of origin 

> length of time in the U.S. 

> place of birth (native vs. foreign bom) 

> age at migration 

> family conposition 

> cortniunity of residence within U.S. (ethnically similar or 

dissimilar) 

> cultural distance from community of origin (frequency of trips to 

place of ujtoringing; degree of assimilation into U.S. culture) 

> changes in social status with migration (relative increase or 

decrease in income; stability and consistency of income) 

> changes in activity patterns 

> changes in social values related to feeding and food (e.g. 

attitudes towards breast feeding; susceptibility to televised food 
conmercials; abandonment of practices perceived as socially 
unacceptable (e.g. chewing bones) 

> practice of geophagia or aitylophagia (clay or starch eating) , 

particularly among pregnant women and children in some cultures 
which may lead to undesirable weight gain (from starch) or other 
nutritional pr(±>lems, such as iron-deficiency anenia (Lackey, 
1983) 

> factors related to enployment of women before/after migration 

> nature of traditional food items and meal patterns 

> similarity between U.S. foods and accustomed food items 

> shifts in meal frequency or times (nuntjer of meals, snacks per day; 

time of day when heaviest meal is taken) 

> shifts in proportions of calories from protein, carbohydrate, and 

fat 

> shifts in protein sources (animal, dairy, fish, poultry, vegetable 

sources) 

> logistical aspects of accustomed methods of food procuranent, 

storage and preparation (daily vs. weekly shopping; requirement 
for long preparation times) 

> types of food substitutions made (non-nutritionally equivalent 

foods within the same food category; manufactured vs. natural 
forms of food) 

> flavor and texture preferences and dislikes related to commonly used 

U.S. foods; 

> perceived social/health attributes of U.S. foods (health related 

food beliefs; types of food viewed as "status" foods) 

> use of vitamin and mineral supplements (discontinued use of 

traditional supplementary foods; U.S. acquired supplement use 
practices) 

> eligibility for and use of food subsidy programs (school lunch, 

government commodities, elderly meal programs, food stairps, WIG) 



131 



The greatest quantity and quality of dietary intake and 
nutritional status data are available for blacks, the minority group 
whose dietary patterns can be expected to differ least from those of 
the general population. With the exception of sub-groups of the black 
population in certain urban areas (e.g., the West Indian commmunity in 
New York City, Haitian refugees in Florida) the population of black 
Americans does not include immigrants from cultures with structurally 
different dietary patterns. CXiltural (rather than socioeconomic) 
differences between diets of blacks and whites are primarily derived 
from differences in regional (southeast ISiited States) and religious 
(more Protestant than Catholic; non-Jewish) influences on food usage 
and preparation and to a lesser extent from slavery-related dietary 
adaptations. In contrast, Asians, Hispanics, and Native Americans may 
retain a carbohydrate-staple structure in their diets which differs 
from the meat staple orientation of the diets of U.S. blacks and 
whites. 

The literature does not give the impression that minority groups 
migrating to the United States are from cultures whose dietary 
patterns are associated with frank vitamin and mineral deficiencies, 
although quantities of food may have been limiting among certain 
migrant groups. In general, large differentials between U.S. minority 
and white population groups can be expected in three areas: 

1) differentials related to the low income status of a high 
proportion of minority individuals and families, including those 
reflected in lowered growth potential of children from chronically 
poor cultures; 

2) differentials related to the excess prevalence of obesity, 
particularly among women in certain minority groups; and 

3) differentials associated with the lower proportions of refined 
carbohydrates and saturated fat, and higher sodium ccaitent, in the 
traditional diets of some minority groups, conpared to the diets 
that minority-group individuals may adapt as they become 
increasingly acculturated. 

In evaluating diets which include unfamiliar foods or food 
coirbinations , care must be taken to differentiate the apparent absence 
of a food from the absence of a nutrient . Only acute nutritional 
defiencies or extreme, long-term maladaptations of formerly adequate 
dietary patterns will result in nutritional risks as they are 
currently defined and assessed. 



132 



3. ASIAN AMERICANS 
3.1 General Dietary Patterns 

The Asian American population consists primarily of Japanese and 
Chinese Americans but also includes Hawaiians, Filipinos, Koreans, 
Vietnamese, Laotians and Cambodians. Among these are individuals and 
families who have been in the Uhited States for several goierations, 
first generation iiradgrants and refugees — the latter in increasing 
numbers. The dietary staple (i.e., the primary source of dietary 
calories) for most of these groups is rice, although varieties used 
and methods of preparation vary. Use of pre-^vashed or unenriched 
refined rice poses risks of low B vitamin and mineral intakes, 
although these risks may be offset by adequate consumption of other 
foods (e.g. pork and fish) . Other starchy foods are also used (wheat 
and rice noodles, tubers). 

Seasoned food mixtures, the secondary source of calories, are the 
major aspect of sub-cultural variation. Intakes of vegetables and 
fruits, fish and shellfish are higher and the animal protein lower 
than in typical U.S. diets. Dairy products are used to a much lesser 
extent among Asians than among the general population. This may pose 
a potential risk for calcium nutriture in Asian American populations 
vrtien the traditional sources of dietary calcium in Asian diets (tofu, 
green leafy vegetables, sardines) are not available or are supplanted 
by otherwise equivalent but lower calcium density foods. "Status 
foods", usually foods whose availability or price have been limiting 
in the traditional diet, may include both nutritious foods (fruits and 
vegetables, ice cream) and low^nutrient density foods (soft drinks, 
candy) (Suitor and Crowley, 1984; United States Department of 
Agriculture, 1980) . The short cooking time of many traditicaial Asian 
foods is favorable to nutrient retention. 

Some characteristic food and food choices of several Asian 
American subgroups are shown in Tables 4 through 7. Adaptations of 
Asian populations to the U.S. diet inevitably involve increased 
proportions of calories from animal protein, fat, and refined sugar 
and decreased calories from coirplex carbohydrates. These changes also 
imply decreased intakes of fiber and increased intakes of cholesterol. 
The higher costs of seafood in the U.S. vs. the home countries of some 
Asian inmigrants may promote shifts toward increased consuption of 
"meat" protein. The evidence available — ^most of which relates to 
Japanese men in Hawaii or California — indicates that dietary changes 
are reflected in greater relative weight and higher coronary heart 
disease irortality among Asians in the United States vs. their 
counterparts in the countries of origin (Tillotson et al., 1973; 
Hank in et al., 1975; Nomura et al., 1978; Kolonel et al., 1981; and 
McGee et al., 1982). 



133 



Table 4: 

Characteristic Chinese Foods and Food Choices 



PROTEIN Meat: pork, beef, organ meats 

FOODS Poultry: chicken, duck 

Fish: white fish, shriiip, lobster, oyster, sardines 

Eggs 

Legumes: soybeans, soybean curd (tofu) , black beans 

Nuts: peanuts, almonds, cashews 



KLIK AND MILK 
PRODUCTS 



Flavored Milk, milk (cooking) , ice cream 



GRAIN PRODUCTS Rice, noodles, white bread, barley, millet 



VEGETABLES BairfxK) shoots, beans: green and yellow, bean sprouts 
bok choy, broccoli, cabbage, carrots, celery, 
Chinese cabbage, com, cucunbers, eggplant, 
greens: collard, Chinese broccoli, mustard, kale, 
spinach, leeks, lettuce, mushrooms, peppers, potato, 
scallions, snow peas, sweet potato, taro, tcanato, 
water chestnuts, white radishes, white turnip, winter 
melon 



FRUITS i^ple, banana, figs, grapes, kumguats, loquats, mango, 
melons, orange, peach, pear, persiimon, pineapple, 
plums, tangerine 



OTHER Soy sauce, sweet and sour sauce, mustard sauce, ginger, 
plum sauce, red bean paste, tea, coffee 



SOURCE: California Department of Health, 1975, Table 11 



134 



Table 5; 

Characteristic Japanese Foods and Food Choices 



PROTEIN 
FOCDS 



Meat: 

Poultry: 

Fish: 



Eggs 
Legumes: 

Nuts: 



beef, pork 

chicken, turkey 

tuna, mackerel, sardines (dried form called 

mezashi) , sea bass, shrinp, abalone, squid, 

octopus 

soi^an curd (tofu) , soybean paste (miso) , 
so^Deans, red beans (azuki) , lima beans 
chestnuts (kuri) 



MIliC AND MILK 
PRODUCTS 



Milk (in cooking) , ice cream, cheese 



GRAIN PRODUCTS 



Rice, rice crackers, noodles (whole wheat 
noodles called soba) , spaghetti, white bread, 
oatmeal, dry cereals (Nisei only) 



VEGETABLES Bantooo shoots, bok choy, broccoli, burdock root, 
cabbage, carrots, cauliflower, celery, cucumbers, 
eggplant, green beans, gourd (Kanpyo) , mushrooms, 
mustard greens, Napa cabbage, peas, peppers, 
radishes (white radish called daikon; pickled white 
radish called takawan) , snow peas, spinach, squash, 
sweet potato, taro (Japanese sweet potato) , tomato, 
turnips, waterchestnuts , yam 



FRUITS i^ple, apricot, banana, cherries, grapefruit, gr^ses, 
lemon, lime, melons, orange, peach, pear, persimmon, 
pineapple, pcmegranate, plums (dried, pickled, plums 
called umeboshi) , stravitoerries , tangerine 



OTHER Soy sauce, nori paste (used to season rice) , bean thread 
(konyaku) , ginger (shoga; dried form called denishoga) , 
tea, coffee 



SCXJRCE: California Department of Health, 1975, Table 12 



135 



Table 6: 

Characteristic Filipino Foods and Food Choices 



PROTEIN Meat: pork, beef, goat, deer, rabbit, variety meats 
FOCOS Poultry: chicken 

Fish: sole, bonito, herring, tuna, mackerel, crab 

mussels, shriiip, squid. 
Eggs 

Legumes: black beans, chick peas, black-eyed peas, 
lentils, mung beans, lima beans, v^ite 
kidney beans 
Nuts: cashews, peanuts, pili nuts 



MILK AND MILK 
PRXUCTS 



Milk: flavored, evaporated 
Cheese: gouda, cheddar 



GRAIN PRODUCTS 



Rice, cooked cereals: farina, oatmeal, 
dry cereals, pastas, rice noodles, wheat 
noodles, macaroni, spaghetti 



VEGETABLES Bairboo shoots, beets, callage, carrots, cauliflower, 
celery, Chinese celery, eggplant, endive, green beans, 
leeks, lettuce, mushrooms, okra, onion, peppers, 
potato, puirpkin, radishes, snow peas, spinach, squash, 
sweet potato, tomato, water chestnuts, watercress, yam 



FRUITS ^^le, banana, grapes, guava, lemon, lime, 

mango, melons, orange, papaya, pear, pineapple, 
plums, pomegranate, rhubarb, stravtoerries, tangerine 



CTTHER Soy sauce, coffee, tea 



SOURCE: California Department of Health, 1975, Table 13 



136 



Table 7: 

Exairples of Some Southeast Asian Foods 



Vietnamese - soup — "pho" — contains rice, noodles, thin slices of 
beef or chicken, bean sprouts, and greens 

- fish an<Vor meat and vegetable dishes 

- clear soup with vegetables anchor meat 

- fish sauce — "nuoc mam" — made from fermentation 

of small fish 

Cambodian - soup with meat and noodles, ancVor rice 

- fermented fish — "prahoc" 

- fish sauce — "tuk-trey" 

- sweets made from palm sugar 

Laotian - fish — "padek" — anchor meat stew with hot peppers 

- sweet or glutinous rice (sticky rice) 

SOURCE: Iftiited States Department of Agriculture, 1980 



Many Asian-American foods and seasonings are very high in 
sodium. Exanples include bean sauces, dried shrimp, dried salted 
fish, pickled vegetables, fish sauce, monosodium glutamate, soy 
sauce, miso, salted eggs (Chew, 1983; Asian/Pacific Islander Task 
Force on High Blood Pressure Education and Control, 1984; Suitor and 
Crowley, 1984) . Thus, dietary acculturation among Asian Americans 
may reduce the level of sodium exposure, and thereby lower 
hypertension risk. Sodium restriction may be particularly beneficial 
for Asian Americans who are hypertensive, but is difficult to 
accomplish within the framework of traditional Asian food practices 
(Chew, 1983; Asian/Pacific Island Task Force of High Blood Pressure 
Education and Control, 1984) . Decreased use of salted fish and 
pickled vegetables may also reduce the risk of certain cancers 
(Kolonel, 1981; Yu, 1981). 



137 



Within the Chinese population, food preparation practices differ 
according to Mandarin (north) , Shanghai (central) , or Cantonese 
(south) influences (Sanjur, 1982). Sanjur (1982) notes two 
ideolgical themes which underlie Chinese food habits; the Fan-Ts'ai 
principle which balances the proportions of starchy staple foods 
(Fan) with meat and vegetable conpaients of the diet (Ts'ai) but 
which considers the Fan coiiponent to be the more indispensible. 
(Sanjur, 1982) . An additional principle in Chinese food practices 
strongly advises against overindulgence (Sanjur, 1982) . This implies 
a relatively low risk of obesity. However, older Chinese may 
associate excess weight with wealth and prosperity, so that obesity, 
if it develops, may be considered acceptable (Asian/Pacific Islander 
Task Force on High Blood Pressure Education and Control, 1984) . 

The Yin-Yang principle is based on beliefs about the inherent 
physiologic significance of foods and, as Sanjur points out, is 
analagous to the hot-cold dichotomy in Hispanic food culture (see 
section 5) . Ludman and Newman (1984) give some exaroples of yin, 
neutral and yang Foods as follows. Yin foods include bland foods, 
boiled foods, cold foods (thermal), some types of fruit, some types 
of fish, pork, most greens, and white foods — including milk. Yang 
foods include fatty meats, hot foods (thermal) , spicy foods, fried 
foods, chicken, beef, and eggs. Noodles, soft rice, sugar, and 
sweets are neutral. 

Health conditions and bod^ organs are also dichotomized according 
to the Yin-Yang principle. Pregnancy and lactation, two periods of 
high nutritional risk, are Yin conditions and would theoretically be 
characterized by de-emphasis of Yang foods (which include high 
protein foods) . However, responses of Chinese-American women to a 
1980 survey of health-related food practices indicates that the 
Yin-Yang principle is not closely followed during pregnancy (Ludman 
and Newman, 1984) . 

Surveys of dietary transitions among Chinese Americans in New 
York City (Sanjur, 1982) , Lincoln, NdDraska (Yang, 1979) , and 
California (Grivetti and Paquette, 1978) indicate that deviation from 
charateristic Chinese food patterns increases with increasing years 
of residence in non-Chinese environments. However, the American foods 
which are accepted or rejected for reasons of ideology or preference 
are difficult to predict and may be very specific to interactions 
between coranpunity of origin and coimunity of residence. In addition, 
patterns of ethnic food use are changing in the home countries of 
recent Asian iitmigrants (Grivetti and Paquette, 1978) . 

Several reports provide specific insights into the dietary 
changes among Vietnamese iirmigrants. Stoner and Grivetti (1978; 
cited in Waldman et al., 1979) noted that westernization of 
traditional food customs in Vietnam began in the 1940 's. Thus, some 
non-traditional dietary practices may precede migration to the U.S. 
These authors noted that Vietnamese refugees in California adopted a 
corrpromise of following U.S. dietary practices during the week and 
returning to more traditional Vietnamese diets during weekends. 

138 



In a study of Vietnamese refugees in Florida, 64% of those 
interviewed reported increases in weight since ccming to the United 
States and had significantly increased their consuirption of milk and 
soft drinks, as well as many other foods (Crane and Green, 1980) . 
Nguyen et al. (1983) ccHipared food habits and preferences of 
Vietnamese children less than 6 years old who had come to the Lftiited 
States more than or less than a year before being interviewed. The 
children who had been in the U.S. longer than one year consumed more 
peanut butter and more sweets (ice cream, milk shakes, and pies) than 
those who had come more recently, but both groups of children 
retained a preference for fruits as snacks. 



3.2 Nutritional Risk 

3.2,1 Infants, Children, and Adolescents 

Stoner and Grivetti (1978; cited in Waldman et al., 1979) 
reported that breast feeding was negatively viewed by Vietnamese 
women in their California study sairple, related to the widespread 
practice of bottle feeding in Vietnam, particularly in urban areas. 
However, a United States Department of Agriculture review (1980) 
states that the majority of Southeast Asian infants are breast fed. 
The CDC pregnancy nutrition surveillance data, which include breast 
feeding data for a subset of postpartum women participating in the 
reporting WIC and Maternal and Child Health Service populations, did 
not include this variable for Asian American women. No other recent 
data on infant feeding patterns among Asian American women were 
identified. Thus, there is not a sufficient basis for making any 
general statement regarding breast vs. bottle feeding patterns or 
possible associated nutritional risks among Asian Americans. 

The CDC-PNSS data indicate an excess prevalence of linear growth 
stunting among the Asian American children in the data base, both in 
coirparison to national standards and in coirparison to the prevalence 
of growth stunting among white children in the CDC population. The 
latter is a coirparison within a population of low-income children 
receiving publicly-supported services. Among Asian-American children 
in the CDC data base who were less than 2 years of age, the 
prevalence of low height for age was 8 to 13% in 1977-78 and 17 to 
20% in 1979-81 compared with the 5% expected (i.e., below the 5th 
percentile) and coirpared with prevalences of approximately 9% among 
white children for the entire 5 year period. Among children ages 2 
to 5 the excess prevalence of growth stunting was more 
pronounced— 21% in 1977 and 33-37% during 1978-81 vs. 5% expected and 
9% in the white children in this data base (Centers for Disease 
Control, 1983) . Neither low weight for height (thinness) nor high 
weight for height (obesity) were in excess prevalence among Asian 
American children in the CDC data for these years. No clear pattern 
of excess risk was evident in the hemoglobin and henatocrit data. 



139 



3.2.2 Adults 

As noted earlier, transition from a traditional Asian to a more 
characteristic U.S. diet may be associated with increases in some 
aspects of chronic disease risk related to higher saturated fat and 
lower complex carbohydrate consunption, and with decreases in 
hypertension risk related to lower sodium consuirption. Some reports 
of inadequate diets among Asian American were noted. Three such 
studies are suirmarized below, primarily as exanples of the types of 
data which have been reported concerning the nutriticaial status of 
Asian Americans. There are insufficient data on which to base any 
general conclusions about undernutrition among Asian American adults. 



Casey and Harrill (1977, cited in Waldman et al., 1979) reported 
on nutrient intakes calculated from 24 hour recalls of Vietnamese 
women relocated to Colorado. Diets of women ages 22 to 48 were below 
the U.S. standards for calcium, iron, and zinc but were otherwise 
adequate. Only protein and ascorbic acid intakes were adequate for 
women ages 51 to 65. No nutrient deficiencies were identified by 
Stoner and Grivetti (1978, cited in Waldman et al., 1979) in their 
study of Vietnamese refugees in California. The majority of 
respondents in that study supplemented their food intakes with 
produce from home vegetable gardens. 

Kim et al. (1984) reported intakes of calcium, selected other 
nutrients, and preferences for calcium-rich foods among 40 elderly 
Koreans in Chicago. The respondents were between the ages of 65 and 
81 and were tested to assure mental coirpetence. Calcium intakes were 
low (below 2/3 of reconinended dietary allowances) for a third of the 
men and two thirds of the women, although slightly higher than 
reported calcium intakes in Korea. Food preference data indicated 
that milk and ice cream were liked by more than 60% of the subjects 
but two thirds of the subjects either disliked or had never tried 
cheese and yogurt. Calcium-rich Korean rich foods were liked by a 
majority of subjects but were infrequently consumed due to cost and 
availability. Other dietary conpcnents thought to have an 
antagonistic effect on calcium nutriture, (e.g., intakes of animal 
protein) were higher than levels in Korea. Thus, the net calcium 
status of the Korean Americans may have been worse than if they had 
been consuming a traditional Korean diet. However, the absence of 
any data demonstrating calcium-related health problems in this 
population limits the significance of these findings. 

Lewis and Glaspy (1975) obtained 3 day food records for 47 
college-educated Filipino women in Los Angeles during 1971. Although 
the diets of many of these women contained Filipino foods, 
significant dietary changes indicative of U.S. dietary patterns were 
reported. The calculated nutrient intakes of these women were quite 
variable; many were below both U.S. and Filipino standards. Intakes 
of some of the women, including two pregnant subjects, were augmented 
by appropriate vitamin and mineral supplemaits. 



140 



3.3 Conclusions 

On the basis of the limited data available on the nutritional 
status of Asian Americans, the most evident area of curroit concern 
is the growth status of Asian-American children. This problem 
presumably relates to the children of recently-migrated Southeast 
Asian refugees rather than to Asian-American children in general. 
The extent to which this excess risk is socioeconomically determined, 
is due to constitutional factors affecting growth potential, derives 
from transitory, migration-related nutritional maladaptations, or is 
a function of the growth standards used to specify risk cannot be 
determined. 

Additional dietary factors which affect the Asian American 
population at large are those related to chronic disease. These 
include elements inherent in certain traditional Asian foods 
(e.g., highly salted and pickled foods) and elemaits acquired as a 
result of acculturation to conventional U.S. eating patterns (e.g., 
saturated fat and refined carbohydrate) . The coronary heart disease 
risk factors are of relatively recent acquisition and may be of a 
lesser concern now than they will be in the future. 



141 



4. NATIVE AMERICANS 



4.1 General Dietary Patterns 

Native Americans include JDOth American Indians and Alaska 
Natives. Very little information vras identified on the diet and 
nutritional status of Alaska Natives. An iirpression of Alaskan 
Eskimo food patterns is provided in the following paragraph. The 
remaining material in this section relates to American Indians. 

The following coirments on dietary patterns of Alaskan Eskimos are 
based on reviews by Gonzalez (1972) and Sanjur (1982) . Artie and 
Subartic food patterns are characterized by extronely high caloric 
intakes. Primitive Eskimo diets were high in protein, of moderate or 
high fat content, limited in carbohydrate content, and were low in 
ascorbic acid (vitamin C) during some seasons. Protein sources 
varied with region and were either sea mammals, fish, or land 
mairmals. Eskimos live at a very low economic level. Many in rural 
areas have unstable incomes. Food supplies are most consistoit 
during the winter months when the freezing tenperatures permit 
long-term storage of excess game. Current diets consist of game and 
store bought items. Important staples include homemade yeast bread, 
fried breads and pilot crackers. Large quantities of unenriched rice 
and pancake mixes are used. Wild berries are widely used; wild 
greens are important in some remote villages. Child feeding 
practices include premastication of infant foods, use of canned and 
evaporated milk, increasing use of corrmercial infant foods; 
widespread use of dry pre-cooked baby cereals, artificially flavored 
powdered drinks, and soft drinks. 

American Indians derive much of their diets from the hunter- 
gatherer diets of early America, Tribal dietary patterns vary 
according to geographic region, cultural, and economic traditions. 
For exanple, diets of Indians living along the northern coast of 
California include a variety of seafood, whereas diets of Indians in 
mountainous areas include more dried meats and fresh water fish 
(California Department of Health) . Dietary patterns of Indians in 
the Southwest may be more similar to those of Mexican Americans than 
to those of Indians from other tribal and regional origins. The 
current diets of the various Indian tribes in the Iftiited States are a 
blend of their traditional foods with the American food culture, 
except for special occasions. For exanple, Indians in the Southwest 
once ate tortillas made from corn, but manufactured wheat tortillas 
have replaced handmade corn tortillas in recent times. In addition, 
factors such as income, the availability of foods, particularly U.S. 
coitmodity foods, stage of acculturation, and place of residence are 
important dietary influences. 



142 



Although there is probably much more heterogeneity than 
homogeneity in American Indian diets overall, most Indian diets 
appear to be characterized by a high carbohydrate, high sodium, and 
high saturated fat content and relatively low content of meat and 
dairy products. Many Indian dishes are made of starch and meat 
combinations and many foods are fried and breaded. Consuirption of 
refined sugar products is high. Cholesterol consumption may he low 
to moderate, depending on the protein sources. 

Exairples of traditional and conteiporary American Indian Foods 
are shown in Table 8. Based on a laboratory analysis of traditional 
foods of Hopi Indians (southwest) Kuhnlein et al, 1979 concluded 
that nutrient needs would be met by the Hopi diet if all the 
quantities of food available were adequate and if all nutrients 
consumed were absorbed. Toma and Curry (1980) calculated the 
nutrient content of seven traditional recipes of South Dakota Indian 
tribes. Five of the seven recipes met the criterion of a nutritious 
food. The food iton of least nutritional value — fried bread — is the 
only one of the seven which has been retained in the customary diet. 



Table 8: 

Examples of Traditional and Contaiporary Foods in the Diets of 

American Indians 



hominy (boiled corn) - pinto beans 

corn balls - mutton, goat 

corn meal - deer 

corn tortillas - salmon 

wheat tortillas - smoked salmon eggs 

oven bread - refried beans 

fried bread - lard and bacon fat back 

piki bread - small game 

cooked cereals - dried meats 

dry cereals - fresh water fish 

wild rice and nuts - seafood 

- dried melon and peach strips 

- wild fruits, berries, and vegetables 

- com, beans, squash, melons 
potatoes, onions, cabbage 



- coffee, tea, milk, soft drinks 

- sweet rolls, cakes, doughnuts 



SOURCES: Gonzalez, 1972; California Department of Health, 1975; 
Alford and Nance, 1976; Kuhnlein et al., 1979; Sanjur, 1982 



143 



Dietary recalls of Eastern Band Cherokee Indian WIC participants 
interviewed in 1978 indicated consunption of several traditional food 
items. Sixteen to 18% of the dietary recalls indicated consuirption 
of cornbread, hominy or boiled corn, pinto beans, and greens; 4 to 8% 
indicated consuirption of fatback, green beans, and fry bread. 
Participation in WIC was reported to influence both the general 
family food consunption patterns and the child feeding practices 
(Slonim et al., 1981) . 

Food intake patterns of Indians on the Standing Itock Reservation 
in North and South Dakota, assessed by Bass and Wakefield (1974) in 
1970, did not revolve around a typical menu or meal except that 
coffee consumption was high and some type of breakfast was usually 
consumed. Consuirption of characteristic white middle class 
foods — including bologna, potato chips and carbcaiated beverages — was 
conmon among those with adequate incomes, while traditional foods 
were served only on special occasions. Coninodity foods were used by 
the majority of families interviewed. Frying and boiling (the 
traditioial method) were used equally. 

Diets of Indians in the Onondaga Nation (upstate New York) are 
characterized by com and legumes as protein sources, high fat-high 
sodium traditional foods and low intakes of meats, eggs and dairy 
products (Sinpson, 1982) . High consumption of refined sugar products 
(e.g., candy, cakes, pies, soda, Kool-Aid) and a predominance of 
fried and breaded foods has been noted among Indians of the St. 
Regis-Mohawk Nation (Rhoades, 1982) . 

4.2 Nutritional Risk 

4.2.1 Infants and Children 

The Centers for Disease Control Pregnancy Nutrition Surveillance 
data for 1981 (Centers for Disease Control, 1983) indicated a high 
prevalence of breast feeding among Native American women at the 
post-partum follow-up visit: 44 percent of Native American vs. 34% of 
white mothers under age 20 were breast feeding; 47% of Native 
American mothers over age 20 were breast feeding, a percentage less 
than the 53% of white mothers but greater than the 30 and 37% of 
Hispanic and black mothers who were breast feeding. Several studies 
in specific Indian populations suf^rt the impression that a large 
proportion of Native American mothers breast feed, and for fairly 
long periods, although a public health nutritionist serving the St. 
Regis-Mohawk nation in New York State noted that women in her service 
population were exceptions in this respect — having a relatively low 
incidence of breast feeding (Rhoades, 1982) . 

A 1978 study showed a beneficial effect of breast vs. bottle 
feeding on the risk of gastroenteritis, as observed among all Pima 
infants under 4 months of age and also among infants of mothers under 
age 20 (Forman et al., 1984). Thirty five percent of White Mountain 
i^che preschool children (Arizona) studied in 1976 were reported to 



144 



have been breast fed — 15% for six months or longer and 25% for at 
least three months. Whereas cow milk was the usual supplementary food 
for breast fed infants, formula fed infants were more likely to be 
supplemaited with evaporated milk (YanochikHDwen et al., 1977) . 
Sixty-six percent of Eastern Band Cherokee Indian women participating 
in a North Carolina WIC program in 1978 reported having breast fed 
their youngest child — 50% for six months or longer — although the 
majority of these women supplemented with bottle feeds from the first 
weeks of life and introduced solid foods within the first three 
months (Slonim et al.rl981). 

From the 1982 CaX-PNSS data, Trowbridge (1983) reported excess 
prevalence of linear growth stunting (12,4% below the 5th percaitile 
of height for age) among Native American one-year-old children but 
not among children less than one year or 2 to 4 years. Cross- 
sectional data for Native American children in the CDC data base 
during 1977 through 1981 indicated a similar pattern but to a lesser 
degree. The prevalence of growth stunting in Native American 
children during 1977-1981 was not markedly greater than the 
prevalence among white children in the CDC data base. Wasting (low 
weight for height) and low hemoglobin were not in excess among Native 
American children (Centers for Disease Control, 1983). 

Yanochik-Owen et al. (1981) reported that three times as many 
White Mountain i^che children surveyed in 1976 were receiving 
vitamin supplements as a similar group of children surved in 1969 
(15% vs. 5%) . Overall energy and nutrient intakes by 24 hour dietary 
recall were higher in the 1976 vs. the 1969 sairple but were below 
critical thresholds for energy, calcium, vitamin C, and iron for 
substantial proportions (16 to 38%) of children in 1976. 
Deficiencies in some areas were probably conpensated by contritxitions 
from supplanents. 

Although clinical signs of prevalent undernutrition were not 
found, anthropometric measures did not reflect the aj^rent 
iirprovonent in dietary quality, possibly because of the short 
intervening time interval. Percoitages of children below the 10th 
percQitile of height for age were nearly three times greater than 
expected in both 1976 and 1969 (39 and 37%) . There were no 
statistically significant differences in height or weight measures 
between 1969 and 1976, although slightly higher thoracic fatfold 
thicknesses were inplied in the 1976 data. The prevalences of low 
hemoglctoin and honatocrit were not high, but the percentages of 
children in the 1976 cohort with low transferrin saturation and serum 
ferritin levels (53% and 22% of children for the two measures, 
respectively) suggested low iron stores — indicative of inadequate 
long term intakes of dietary iron. 

Uhdemutrition — defined by clinical signs, hanoglobin levels, 
height for age and urinary hydroxyproline to creatinine ratio — was 
not prevalent among the Chippewa Head-Start children in Wisconsin 
studied by Homer et al. (1977) . Protein intakes were well above the 
RDA (172%) ; dietary iron intakes were adequate (80% of the RDA) . 

145 



The prevalence of lactase deficiency among 156 Chippewa children 
ages 5-17 and adult volunteers on the Leech Lake Reservation in 
northern Minnesota varied from 62 to 72% and did not vary within the 
age ranges studied. Lactase deficiency was almost universal among 
full-heritage Chippewa and was least prevalent (around 33%) among 
those with less than 50% Indian heritage. However, milk intolerance 
and milk consuirption were only slightly lower among the lactase- 
deficient vs. the lactase-normal subjects studied. Only 10% of 
children reported milk intolerance (Newccaner et al., 1977). 

An excess prevalence of overweight was observed among Native 
American children in the CDC-PNSS in 1982 (13.2, 12.0, and 10.4% of 
children ages 1,2, and 4, respectively, above the 95th percentile of 
weight for height) (Trovfcridge, 1983) . Percentages of overweight 
children less than one year old and 3 years old were sc»newhat greater 
than expected as well (7 and 8.2% respectively) . Trends during 1977 
to 1981 indicated a prevalence of high weight for height among Native 
American children which was consistently higher than expected and 
higher than the prevalence for white children in the sairple, 
particularly in the 2 to 5 (vs. less than 2) year age group. 

Excess prevalence of obesity was reported among Chippewa 
preschool children in Wisconsin (Homer et al., 1977), Thirty 
percent of boys were above the 90th percentile of weight for height; 
19% of girls and 37% of boys were above 110% of the NCHS reference 
standard of weight. Percentages of children with skinfold 
thicknesses above the 90th percentile were 35% and 37% for girls and 
boys, respectively. 

4.2.2 Adults 

In a study of pregnant and lactating Navajo women, Butte and 
colleagues (1981) found the median nutrient intakes of these women to 
be less than 60% of the RDA for calcium, magnesium, zinc, copper, 
vitamins A, D, E, B6, biotin, and folacin. Anonia was present in 
15-20% of the women. The authors note that the total energy intake 
of these women was only 74% of the PDA, which is inadequate for 
pregnancy and lactation. However, it should be noted that 60% of 
1200 mg/day, the calcium RDA for pregnant and lactating women, is 720 
mg/day, nearly equal to the 800 mg/day RDA for non-pregnant women. 
Their diets may actually provide adequate calcium and other nutrients 
for non-pregnant women. 

A 1970 survey indicated that dietary intakes of Indian people 
(primarily Sioux) on the Standing Rock Reservation in North and South 
Dakota were adequate (2 67% of the RDA's) or in excess of the RDA's 
for most nutrients. Calcium intakes and iron intakes of women of 
reproductive age were substantially below recoirmended levels and more 
than half of the women interviewed reported calcium and vitamin A 
consuirption below 50% of the RDA (Bass and Wakefield, 1974) . Actual 
iron intakes may have been higher than calculated due to the 
customary use of iron cooking utensils. 



146 



Mthough oiDesity was reportedly uncoirmon among Native Americans 
before the 1940 's (West, 1974), documentation of the excess 
prevalence of obesity among Native Americans in all parts of the 
country is now frequently encountered in the literature. Johnston et 
al. (1978) have observed that urban Native Americans are at higher 
risk for obesity than those who remain on reservations. Gillum et 
al. (1980; 1984) have reported an excess prevalence of obesity (vs. 
white coirparison populations) among Native American (primarily 
Chippewa) school children and adults in Minnesota in 1978 and 
1980-81. 

Studies dating back to the 1960 's have noted an excess of ctoesity 
among various Indian populations — White Mountain inches (Clifford et 
al., 1963), Navajo (Fulmer and Roberts, 1963; DeStefano et al., 1979, 
Chase-the-Bear et al., 1979), Seminoles in CStlahoma (Mayberry and 
Lindeman, 1963) , Sioux Indians in North and South Dakota (Bass and 
Wakefield, 1974) and Seneca Indians in New York State (Do^lin et 
al., 1969; Judkins, 1978). Ctoesity among the Pima Indians has 
received considerable attention in conjunction with the extremely 
high prevalence of adult onset diabetes mellitus and the apparently 
lower health risk among Pimas at levels of obesity usually considered 
severe in the general U.S. population (Pettitt et al., 1982). A 
higher than average prevalence of adult onset diabetes has also been 
reported for Indian tribes other than Pimas (Mayberry and Lindeman, 
1963; Do^lin et al., 1969; Judkins, 1978; Gillum et al., 1984) 

Alcohol use is high in many Native American comnimities . 
According to a 1960-65 study by Sievers (1968) , heavy alcohol use was 
more common Among Southwestern Indians than among Indians outside of 
the Southwest but was still higher in both groups of Indians than in 
a white coiiparison population. Striirbu and Sims (1974) reported that 
American Indian high school students in a Georgia survey reported 
levels of alcohol consuirption higher than that of students in most 
other ethnic groups. Forty percent of Navajo men under age 40 
reported using alcohol in a 1977 study, compared with only 22% of the 
men over 40 — a possible sign that alcohol use is increasing among the 
younger Navajo men (DeStefano, 1979) . Alcohol intake was reportedly 
low among the Chippewa Indians studied by Gillum et al. (1984) in 
Minneapolis in 1980. 

4.3 Conclusions 

In summary, even with the limited amount of data available, the 
need for attention to the prctolems of <±)esity and obesity-related 
diabetes in Indian populations is evident. Other dietary risk factors 
for ischemic heart disease may become important for this population 
in the future. In addition, the excess prevalence of both growth 
stunting and obesity among low^ income Native American children may 
indicate poor dietary quality in some elements critical for growth. 
Alcohol-related nutritional risks are also worthy of attention in 
this population. 



147 



5. HISPANIC AMERICANS 

5.1 General Di etary Patterns 

Among the Spanish-speaking U.S. minority groups, the largest 
proportion are Mexican Americans. Puerto Ricans are the second 
largest group. There are substantial numbers of Cuban Americans as 
well. Studies of Hispanic Americans may refer to one or several of 
the following groups: migrant workers, recent immigrants from Mexico, 
long term U.S. residents of Mexican descent, Puerto Ricans in Puerto 
Rico, Puerto Ricans on the U.S. mainland, Cuban immigrants and their 
descendants, or individuals of other Spanish or Latin American 
origins (i.e., with Spanish surnames). The socioeonomic status and 
degree of acculturation vary greatly among these groups, as do the 
referent cultures. The information in this section relates primarily 
to Mexican Americans and Puerto Ricans, due to the relatively few 
available data on Cuban Americans. 

Some characteristic Mexican American and Puerto Rican foods are 
shown in Tables 9 and 10. Mexican American and Puerto Rican diets 
are similar in some respects, but there are underlying differences 
associated with the Southwestern vs. Caribbean origins of these two 
Hispanic cultures. Both Mexican American and Puerto Rican diets have 
carbohydrate staples (tortillas anchor rice) . Consuiiption of green 
leafy vegetables is limited. Vegetable protein sources are given 
more eitphasis in traditional Hispanic diets than are animal 
proteins. Consumption of milk and dairy products is low in 
traditional Hispanic diets, but is not absent. For exanple, 
characteristic foods include ice cream, custard (flan) , pudding 
(e.g., arroz con leche) , and coffee with milk (cafe con leche, 
oatmeal cooked with milk (avena) (California Department of Health, 
1975; Sanjur, 1982). The more conventional uses of dairy products 
are increasing in association with participation in subsidized food 
programs such as WIC and School Lunch (Yanochik-Owen and White, 1977; 
Lid^erman, 1979) . 

The food selections and preparation practices in Mexican 
American and Puerto Rican diets can potentially provide all essential 
nutrients. At one time the food patterns of New York City-based 
Puerto Ricans were publicly criticized as being inherently deficient 
in folacin and therefore predisposing to a widespread prevalence of 
megaloblastic anemia (Sanjur, 1982) . A study of Puerto Rican foods 
and food preparation methods indicated that the folacin in the 
typical diet is adequate (Parker and Bowering, 1976) . However, the 
Hispanic diet is bulky and consists of more moderate nutrient density 
than high-nutrient-density foods. Adequate quantities of nutrients 
will be obtained if a sufficient quantity of foods is consumed. If 
food quantity is limited, e.g. by income, then intakes of essential 
nutrients may be marginal or low. 



148 



Table 9: 

Characteristic Mexican-American Poods and Food Choices 



PROTEIN Meat: beef, pork, lamb, tripe, sausage (chorizo) , 

POODS bologna, bacon 

Poultry: chicken 

Eggs 

Legumes: pinto beans, pink beans, garbonzo beans, lentils 

Nuts: peanuts, peanut butter 



miK AND MILK 
PRODUCTS 



Milk: fluid, flavored, evaporated, condensed 
Cheese: American, Monterey jack. Hoop 
Ice cream 



GRAIN PRODUCTS 



Rice, tortillas: com, flour, oatmeal, 

dry cereals: cornflakes, sugar coated; noodles 

spaghetti, white bread, sweet bread (pan dulce) 



VEGEEABLES 




FRUITS ^ple, apricots, banana, guava, lemon, mango, melons, 
orange, peach, prickly pear cactus fruit (tuna) , 
zapote (or sapote) 



OTHER Salsa (tomato-pej^r onion relish) , chili sauce, guacaraole, 
lard (manteca) , pork cracklings, fruit drinks, Kool-aid, 
carbonated beverages, beer, coffee 



SOURCE: California Department of Health, 1975, T&ble 9 



Mexican American and Puerto Rican diets are gaierally favorable 
fran a chronic disease risk perspective. Fiber content is high. 
Animal fat content is an important conponoit of Hispanic diets but 
its proportional contribution to calories is less than in typical 
U.S. diets. The high carbohydrate content does pose a potential risk 
of excess caloric intake if the meal pattern shifts f ran the 
tradition of taking a heavy meal at midday to the U.S. practice of 
having the heaviest meal in the evening — ^particularly if the heavy 
evening meal is added to, rather than substituted for, the midday 
meal. 



149 



Table 10: 

Characteristic Puerto Rican Foods and Food Choices 



PROTEIN Meat: beef and pork more often than other meats; 
POCDS pig intestines, either fried (cuchifritos) or 

stewed (sancocho) ; organ meats; blood sausage 

(chorizo) ; ham butts, sausage to flavor certain 

dishes 
Poultry: chicken 

Fish: in limited amounts; salt codfish is most coninon 
Eggs: fried and in cooking 
Legumes: white, kidney, pink, beans; pigeon peas, 

chick peas 



MILK AND MILK 
PRODUCTS 



Milk: in limited amounts, e.g. in coffee; in 

oatmeal 
Cheese: native white cheese (resembling farmer 
cheese; gouda cheese, American cheese 



GRAIN PRODUCTS 



Rice, French bread, rolls, crackers, other bread; 
breakfast cereals: oatmeal (avena) , cornmeal, 
and cornflakes; farina 



VEGETABLES Yautia, name, plantain, punpkin, carrots, 
sweet potatoes, tomatoes, onions, lettuce, 
cabbage, potatoes; malanga, yuca (root vegetables); 
white and yellow yams, eggplant 



FRUITS 



bananas, pineapples, guava, oranges, papaya, 
acerolas (West Indian cherry) , mango 



OTHER sof rito (sauce with green pepper, tomato, garlic, lard) ; 
lard and salt pork; olive oil and other vegetable oils; 
sweetened beverages, cakes, pies; guava, orange, and mango 
pastes, boiled papaya preserves; fruit cocktails, canned 
fruits; pear, peach, and apricot nectars; canned soups; 
black malt beer; annate seeds (used for coloring) 

SOURCE: New York City Department of Health, 1976a; 1981; 
Sanjur, 1982 



150 



Sausages and some of the sauces in Mexican American and Puerto 
Rican diets are high in sodium (Suitor and Crowley, 1984) . To the 
extent that these high sodium foods or seasonings are frequently 
consumed, sodium-related risks in the traditional Hispanic diet may 
be similar to those in the general U.S. population. Higher than 
average sodium consunption among Mexican Americans was suggested by 
Kerr et al. (1982) . These authors found that table salt sales in 
Hispanic neighborhoods in Houston, Texas were twice as high as those 
in white neighborhoods. 

Dietary practices of Hispanic Americans may be influenced by 
health-related food beliefs derived frcMii the hot-cold dichotomy of 
body organs, diseases, and other aspects of life. According to 
Sanjur (1982) the basic pronises of this ideology are that a person's 
normal, healthy state is teirperate and that in order to maintain this 
healthy state the individual must balance heat and cold. The 
designations assigned to particular food and beverage itans as hot, 
cold, or teirperate may vary within Hispanic subcultures. Galli 
(1975) gives the following categorization for Puerto Ricans; cool 
foods — whole milk, barley water, chicken, raisins, most fruits, and 
honey; cold foods — avocado, bananas, coccoiut, lima beans, and sugar 
cane; hot foods — evaporated milk, vitamins, iron supplements, calcium 
pills. 

The effect of hot-cold food categorizations on weaning practices 
and on the food intakes of pregnant and lactating women is an 
important nutritional risk consideration. According to Galli (1975) , 
avoidance of cold foods is advised for 40 days after birth. 
Similary, Sanjur (1982) cites studies circa 1960 indicating that some 
Mexican-American pregnant women excluded all fruits and vegetables 
for 30 to 40 days or for the entire lactation period. The current 
influence of "hot-cold" food classifications on food intake among 
Hispanic women cannot be estimated. The iitpression given by a 1979 
report of Lieberman is that such beliefs have been integrated with or 
replaced by more modem "scientific" beliefs; however, that study 
population was apparently characterized by more rapid acculturation 
than is usual among mainland Puerto Rican conmunities. 

5.2 Nutritional Risk 

5.2.1, Infancy and Childhood 

Several relatively recent studies of infant feeding patterns 
among Hispanic Americans were identified. These studies indicate 
considerable differences between Hispanics and white or black 
conparison groups in the incidence and duration of breast feeding, in 
the determinants of a decision to breast feed, in the types of 
supplementary or substitute milks used in bottle feeding, in the 
timing of introduction of solid foods (which is related to breast 
feeding patterns) , and in types of solid foods used (types of 
conmercial foods used and relative proportions of coitinercial and 
table foods given at a particular age) . 



151 



Several studies have indicated a low incidence of breast feeding 
among Hispanic-American women. Among a subset of postpartum wcanen in 
the CDC-PNSS data base in 1981, the incidence of breast feeding was 
21% among women under age 20 and 30% among wonen over age 20, 
coicpared with 34 and 53% among whites, A stu(^ of infant feeding 
practices among 28 Cuban, 28 Puerto Rican and 20 Anglo families in a 
Dade County, Florida, Maternal and Infant Care program reported that 
the incidence of breast feeding was higher among Anglos than among 
Latin women (Bryant, 1982) . Only 27 of 76 wcaonen breast fed at all; 
35% of Anglos breast fed, 10% of Puerto Ricans, and 12% of Cubans, 
Low incidence of actual or planned breast feeding among Mexican 
American women has also been reported by Seger et al, (1979) , Magnus 
(1983) , Shiith (1982) , and Rassin et al, (1984) , The trend data for 
1971-79 reported by Shiith et al. (1982) indicated that the increased 
breast feeding incidence seen among Anglo women is not evident among 
Mexican Americans. The low breast feeding incidence among Hispanic 
American women is influenced, but not entirely explained, by 
socioeconomic factors. 

Bryant (1982) reported that the Cuban and Puerto Rican women 
studied in Florida changed to bottle feeding after 2-6 weeks — using 
condoised, evaporated, and whole cow's milk in addition to conmercial 
formula. Fifty percent of both Hispanic and Anglo wanen added 
sucrose to the fomula. Ferris et al. (1978b) reported that in a 
sanple of infants in Western Massachusetts, intakes of iron fortified 
formulas, cereal, fruits and vegetables were lower among Hispanic 
infants than among black or white infants. Hispanic infants were 
more likely than black and white infants to be fed fruit juice, as 
well as vegetable-meat mixtures. 

Among Puerto Ricans in East Harlem before the introduction of 
the WIC program, Bowering and coworkers (1978) found that Puerto 
Ricans used whole cow's milk at 3 months while blacks used formula. 
Neither group breast fed their infants. Introduction of solid foods 
was early (at 1 month) . Between 9-15 months, Puerto Ricans used baby 
foods for 30% of oiergy intake, v^ile blacks used baby foods to 
provide only 5% of energy intake (table food provided most of the 
snacks) . Puerto Ricans used more juices than blacks. Meat was used 
by 6% of Puerto Ricans at 3 months; no blacks used meat at 3 months. 
More Puerto Ricans than blacks used cereal. Vitamin supplements were 
given to 75% of infants in this sanple; iron suEplonents were given 
to 90%. 

These studies suggest a less than optimal prevalence of the 
currently recomnaended infant feeding practices in Hispanic 
communities. The effects of expanded WIC program availability on 
infant feeding patterns is undoubtedly positive, but some problems in 
the area of infant feeding may persist. 



152 



CDC-PNSS data for 1977-81 and 1982 (Centers for Disease Control, 
1983; Trover idge, 1983) for children less than 2 years of age do not 
show an excess prevalence of low weight for height or low height for 
age among Hispanic children, coirpared with reference standards or 
with white children in the data base. However, an excess prevalence 
of low height for age (linear growth stunting) is inplied in the data 
for children over 2 years old; for exanple, the prevalence among 
Hispanic children was 16% (vs. 5% expected and 6% among v^ite 
children) in 1982. Findings of an excess prevalence of growth 
stunting among Hispanic Arnerican children have also been reported in 
several other studies (Yanochik-Owen and White, 1977; Lowenstein, 
1981; Dewey, 1983; Alvarez et al., 1984). The Alvarez et al. (1984) 
data for Hispanic children in an inner-city neighborhood health 
center population during 1978 also indicated a 13% prevalence of 
acute undernutrition (primarily moderate rather than severe levels of 
underweight, according to the ratio of observed weight to the 
age-appropriate NCHS median) . The prevalence of short stature was 
greater among inmigrant than among U.S. bom Hispanic children. 

A few additional sources have reported low iron intakes, low blood 
iron status, or low intakes of other nutrients among Hispanic 
children or adolescents (Yanochik-Owen and White, 1977; Haider and 
Wheeler, 1980; Lowenstein, 1981) . The 1977-1981 CDC-PNSS hemoglobin 
data did not indicate an excess prevalence of low values among 
Hispanic children (Centers for Disease Control, 1983) . Data from the 
National School Lunch Program evaluation did not find major 
differences between the 24 hour dietary intakes of Hispanic vs. white 
children, although it was noted that Hispanic children were more 
likely than white children to have energy and calcium intakes below 
the RDA (Vermeersch et al., 1984). Overall, these data do not give 
an inpression of any clear pattern of dietary deficiency. The data 
on stunting are much more consistent. 

An excess prevalence of overweight has also t»een a common 
finding among Hispanic American children (Yanochik-Owen, 1977; 
Lowenstein, 1981; Centers for Disease Control, 1983; Trovferidge, 
1983) . According to CDC-PNSS reports for 1977-1982, the prevalence 
of high weight for height among Hispanic infants was somewhat greater 
than expected but was similar to that among white children in the 
data hjase. Among older Hispanic children the prevalence of high 
weight for height was higher than expected and higher than for the 
white children (Centers for Disease Control, 1983; Trovtoridge, 1983) . 

5.2.2 Adults 

Cardenas et al. (1976) reported that intakes of meats, milk, 
fruits and vegetables were lower for primigravid Mexican American 
women than "average" Anglo women. Thirty-nine percent of the Mexican 
American women were overweight or obese at the first visit. Hunt et 
al. (1979) reported that 85% of low income Mexican American pregnant 
women had reported intakes below 2/3 of the 20 mg/day EDA for zinc; 



153 



protein intakes less than 2/3 of the RDA were reported by 30%. 
Haider and Wheeler (1979) compared nutriait intakes of Hispanic and 
black mothers in Brooklyn. Intakes of fiber, calcium, 0x)sphorous, 
iron and riboflavin were higher among Hispanic than among black women 
and the diets of the Hispanic women had greater variety than those of 
black women. However, diets of the Hispanic wcMien averaged caily 2/3 
of the PDA for energy, were higher in protein and fat than in 
carbohydrate content, and low in vitamin A. 

Recent reports of the practice of geophagia were not identified. 
Sixteen percent of Mexican American migrant women studied by Larson 
et al. (1974) in 1970-72 reported eating dirt or clay during 
pregnancy. 

Based on analyses with data from the National Food Consuitption 
Survey, Windham et al.(1981) reported that nutrient density is 
relatively caisistent across sex and age for all U.S. individuals 
over age 4. In a later study (1983a) these authors examined the 
effect of age, ethnicity, and socioeconomic variables, on the 
nutrient density of diets reported in the WCS, Multivariate 
procedures were used to adjust for differences in age, sex, weight 
and height distributions among racial groups. Since the amount of 
seasonal variation in reported dietary intakes was low, data from the 
spring quarter of the survey were used. Intakes of calcium, 
magnesium, vitamin A and thiamin per 1,000 kcal differed 
significantly by race. Average calcium density of diets was lowest 
among Hispanic and black groups, averaging 30-40 mg per 1,000 kcal 
less than whites. Diets of Hispanic Americans were of lower vitamin 
A density, by 800 to 1,600 lU per 1,000 kcal, than whites or blacks. 
The average thiamin density of Hispanic diets was higher than among 
other ethnic groups. 

Nutrients for which dietary density did not differ by racial or 
socioeconomic factors in the model tested were protein, iron, 
phosphorous, riboflavin, and vitamin B12 (per 1000 kcal) . In analyses 
conparing nutrient density with recommended allowances (i.e., using 
the Index of Nutritional Quality (INQ)), Windham et al. (1983b) found 
that the only substandard nutrient intakes which were significantly 
related to race were vitamin A intakes among Hispanic Americans. 
Average INQ values were 0.95 (where l=adequate) for Hispanic 
Americans vs. 1.39 and 1.79 for v^ites and blacks, respectively. 

Several authors have raised concern over the prevalence of 
chronic-disease related ovemutrition among Mexican Americans. Serum 
cholesterol levels are reportedly similar or lower among Hispanic vs. 
Anglo Americans (Yanochik-Owen and White, 1977; Williams et al., 
1979; Stern et al., 1982). Obesity and diabetes are in excess 
prevalence among Mexican Americans. Haider and Wheeler (1979) noted 
that Hispanic mothers in Brooklyn were overweight and had 
above-standard mean triceps skinfolds values. From the San Antonio 
Heart Study, Stem et al., (1982) reported an excess of overweight 
among both low-income and suburban Mexican Americans compared to 

154 



Anglo Americans of comparable socioeconomic status. Mexican 
Americans are 2 to 4 tiroes more likely than Anglos to be obese (Stern 
et al., 1983). Hanis et al. (1983) reported that 50% of Mexican 
Americans in Starr County, Texas v?ere either diabetic (adult onset) 
or had a first degree diabetic relative. The excess of obesity does 
not entirely explain the excess prevalence of diabetes (Stern et al., 
1983) . 

Findings related to dietary aspects of coronary heart disease 
risk have been reported from the Puerto Rican Heart Study. Relative 
to the men in the Framingham Study, Puerto Rican men consume fewer 
calories, less cholesterol and proportionately less fat, protein, and 
alcohol calories (Garcia-Palmieri, 1977) . Within Puerto Rico, 
significant urban-rural differences have been reported for relative 
weight and serum lipids (rural=lower) (Garcia-Palmieri, 1977). A 
very small proportion of the lipid differences {^proxiamtely 2.5%) 
could be statistically explained by urban-rural differences in 
diet. 



5.3 Conclusions 

Data from the Hispanic HANES (HHANES) will permit a detailed 
description of dietary patterns and nutritional status among Mexican, 
Puerto Rican and Cuban Americans. Thus, although the dietary and 
nutritional status picture for Hispanics is based on limited 
information at this writing, a much more substantial basis for 
nutritional risk inferences will be available in the near future. 

The evidence reviewed is consistent in demonstrating a 
disproportionate risk of both stunting and obesity among Hispanic 
children. Although the problems of stunting are undoubtedly confined 
to populations of lower socioeconomic status, and to immigrant and 
migrant worker subgroups, the prdDlem of obesity in Mexican Americans 
is seen across socioeconomic levels. Areas of special dietary and 
nutritional concern which can be pinpointed in the Hispanic 
population include the low incidence of breast feeding, some 
associated infant feeding practices which are not in keeping with 
current pediatric guidelines, a tendency toward suboptimal intakes of 
vitamin A, and the excess prevalences of obesity and diabetes. Other 
areas of nutritional concern which apply to the gaieral population 
should not be overlooked where Hispanics are concerned. The dietary 
practices of Hispanic Americans are more favorable to low corcmary 
heart disease risks than diets of Anglo Americans, but the 
differences may decrease as degree of acculturation increases. In 
addition, the nutritional situation of Hispanic refugees should be 
given continued attention. A high prevalence of nutritional problems 
among a group of Cuban refugees in Florida has been noted (Gordon, 
1982) . 



155 



6. BLACK AMERICANS 

6.1 General di etary patterns 

If the black Ainerican population is to be subdivided on the basis 
of dietary pattern variations, the subgroups of possible interest are 
southern- vs. non-southern-bom blacks, urban vs. rural, native-born 
vs. imnigrant, and Christian vs. Muslim blacks (anong native-bom 
blacks) . Fifty-three percent of the black population lives in the 
South (U.S. Bureau of the Census, 1984) and a substantial proportion 
of blacks in other regions are southern-bom) . Black inmigrants 
include persons from Caribbean islands and African countries. 
However, S^^anish-speaking black inmigrants are most often classified 
as Hispanic. The material in this section relates primarily to 
American-bom blacks. Some connents on Caribbean dietary patterns 
have been included. Dietary patterns based on Muslim dietary laws 
are not specifically addresssed. 

It should be stated at the outset that there is no black American 
dietary pattern in the same sense that there are Asian, Hispanic, or 
Native American dietary patterns. Structural dietary influences, if 
any, of the African origins of the American black population are not 
evident. The "southern" aspects of black diets are ethnic in the 
sense that most black Americans live or have origins in the U.S. 
Southeast. Sanjur (1982) notes the probability of more "eating out" 
among black Americans due to the high proportion of blacks who reside 
in urban areas. Only 14.7% of blacks were classified as "rural" in 
the 1980 census (U.S. Bureau of the Census, 1984) . The regional, 
residence area, and socioeconcnnic status conparability of black and 
white survey saitples influences interpretation of differences in 
black and white food consuirption patterns. For exanple, black-white 
food pattern differences in data for the overall U.S. population may 
not be evident among blacks and whites in the South or in rural 
areas. Diets of blacks and whites living in the same regional area 
and of the same socioeconomic level tend to be more similar to each 
other than to diets of whites and blacks in different regional areas, 
and/ot at different socioeconomic levels. 

Some chararacteristic black foods and food choices are shown in 
Table 11. The unique cultural, rather than regicxial or socioeonomic, 
influence on black dietary patterns relates to the types of foods 
available to blacks during slavery. Sanjur 's review (1982) points 
out the different dietary patterns of field vs. house slaves. Slaves 
who worked in the fields developed meals requiring minimal 
preparation which were suitable for preparation in large quantities. 
Characteristic foods made use of cuts of pork which were inexpensive 
and undesirable to whites (e.g., tail, feet, chitterlings, ears) 
(Sanjur, 1982) . I3x>se who worked as house servants ate diets more 
like those of the planation owners. 



156 



Meal size and frequency were influenced by patterns associated 
with farming (i.e., heavy breakfast and large midday raeal) . Pood 
preparation methods and flavoring principles characteristic of soul 
food include frying, smothering and barbecuing meats, use of black 
peeper and hot sauce, ham hocks, and salt pork as essential 
flavorings. Greens prepared in traditional style are boiled for long 
periods; however, nutrients which leach into cooking water are 
consumed with the greens as "pot liquor". 



Table 11: 

Characteristic Black Foods and Food Choices 



PROTEIN 
POC»S 



Meat: 

Poultry: 
Fish: 

Eggs 
Legumes: 

Nuts: 



beef, pork and ham, sausage, pig's feet, ears, 
etc., bacon, luncheon meat, organ meats 

chicken, turkey 

catfish, perch, red snapper, tuna, salmon, 
sardines, shrimp 

kidney beans, red beans, pinto beans, 

black-eyed peas 
peanuts, peanut butter 



MILK AND mm 

PRODOCTS 



Milk: fluid, evaporated in coffee, buttermilk 
Cheese: Cheddar, cottage 
Ice cream 



GRAIN PRCDUCPS Rice, combread, hominy grits, biscuits, white 
bread, dry cereal, cooked cereal, macaroni, 
spaghetti, crackers 



VEGEJEABLES Broccoli, cabbage, carrots, com, green beans, 

greens: mustard, collard, kale, spinach, turnips, 
etc.; lima beans, okra, peas, potato, pumpkin, 
sweet potato, tomato, yam 



FRUITS ^^le, banana, grapefruit, grapes, nectarine, 
orange, plums, tangerines, watermelon 



OTHER Salt pork, fruit drinks, carbonated beverages, 
gravies 

SOURCE: California Department of Health, 1975, Table 10 



157 



Carry-overs of some slavery-related and southern dietary 
traditions are designated by the general term "soul food". No 
studies of consumption frequencies for soul food, as such, were 
identified. Soul food-related preferences and consuitption patterns 
persist to some degree (Wyant and Meiselman, 1979; Cronin et al., 
1982; Gite and Perry, 1983) . Certain items such as com bread, grits, 
and greens, which are non-controversial and which lend thanselves to 
daily meal patterns are probably consumed according to habit and 
individual preference. Items such as chitterlings are probably 
reserved for special occasions. The consuitption of pork 
products — either in general or with specific respect to "waste" cuts 
given to slaves — is somewhat controversial among some black 
Americans, particularly in the younger generations. The controversy 
apparently steons from a substantial diffusion of Maslim dietary 
principles among blacks in urban inner cities, intermingled with a 
deliberate rejection of cultural traditions which are associated with 
slavery. 

Relevant positive* features of traditional black food choices 
include the rich sources of vitamin A (yellow and dark leafy green 
vegetables) and thiamin (pork) , and the relatively high fish and 
poultry content. Unfavorable features include the high sodium caitent 
of many items, the extensive use of frying, the use of less 
nutritious pork cuts, the overcooking of vegetables, and a tendency 
toward large, heavy meals. There does not seen to be an 
ideologically-based food hierarchy in traditional black diets. 
However, certain health-related food beliefs may persist, 
particularly among older blacks who were raised in the South 
(Comely, 1963; Snow, 1976). 

As described by Snow (1976) , southern medical folklore includes 
an idea that some foods will "make the blood go up or bring it down" 
in volume, leading to "low or high blood". Foods thought to cause 
"high blood" or to build up "low blood" include beets, grape juice, 
red wine, liver, red meat, and other red foods. Foods thought to 
thin down "high blood" include lemon juice, vinegar, epsom salts, 
sour oranges, and brine from pickes or olives. Pregnancy was 
considered to be a condition during which a natural tendency toward 
"low blood" was assumed. Thus these beliefs favored the intake of 
several nutrient dense foods by pregnant women. However, failure to 
distinguish "low and high blood" from low or high blood pressure, may 
have led to increased intakes of olive or pickle juice, which are 



*With respect to vitamin A, some limited evidence asscx;iating the 
excess risk of prostate canc:er in black men with the level of vitamin 
A consunption has been noted (Graham et al., 1983; Ehterline, 1984) . 
The postulated prostate cancer association with vitamin A consuirption 
is such that risk is higher at higher levels of intake. In contrast, 
there is a considerable amount of evidence that high vitamin A 
intakes are associated with lower lung cancer risk. 



158 



very high in sodium, to "thin the blood" of those with high blood 
pressure. This is in spite of an equally longstanding belief among 
blacks that high blood pressure is aggravated by salt (Ward, 1982) . 
The extent to which ttese beliefs affect current dietary practices of 
southern blacks is unknown. 

Data on food consuroption patterns or food expenditures of blacks 
continue to reflect the traditional black American dietary errptoses, 
i.e., relatively high meat diets, but weighted more to fish, poultry, 
eggs, and pork than beef; high vitamin A consunption, high salt 
consumption; somewhat lower intakes of dairy products and desserts. 
Bureau of Labor Statistics Consumer Expenditure Diary Survey data for 
1972-1974 indicated that, at equivalent income levels, blacks spent 
more on beef, pork, poultry, fish and seafood than vrtiite households; 
expenditures for cereal and bakery products, dairy products, sugar 
and other sweets were less than those of white households (Salathe et 
al., 1979). Black households in urban areas of the Northeast and 
North Central regions spent more on non-alcoholic beverages than 
whites; expenditures on non-alcoholic beverages were lower among 
blacks than whites in other regions of the country. Market Research 
Corporation of America data on dairy product purchases of southern 
households in 1972-74 indicated that per capita purchases of cheese, 
fluid milk (except buttermilk) , and dairy products overall were lower 
in black than in white households. Congruent with these findings, 
the levels of calories, protein, and calcium obtained from dairy 
products were less among blacks than among whites by ^proximately 
50% (Kreidler et al., 1980). 

A 1973 telephone survey of beverage consuirption patterns among 
approximately 8500 residents of upstate New York and New York City 
(Cook et al., 1975) indicated susbtantially higher water consunption 
and substantially lower coffee and tea consunption among blacks than 
whites. Smaller differences in consunption patterns for milk (blacks 
less) , fruit juice or drinks and soft drinks (blacks more) were seen 
between blacks and whites in certain sex-age groups. 

Results of an analysis of data from the National Food Consunption 
Survey (1977-78) reflect some of the expected regional/racial 
differences in dietary pattern eiphases (Cronin et al . , 1982) . A 
larger proportion of blacks and southerners than vrtiites and 
non-southerners reported consumption of bacon and salt pork; more 
blacks than whites reported c<»isunption of dark green vegetables, 
dried beans and peas, and poultry; fewer blacks than v^ites reported 
using milk, yogurt, or cheese and beef. A comparison of black and 
white food frequency data from NEIANES I reflected higher intakes of 
meat, eggs, vitamin A, and salty snack foods among blacks than among 
whites but did not indicate substantial racial differences in the 
percentages of persons using milk; use of desserts, fruits and 
vegetables was lower among blacks than among v^ites (Villa Dresser et 
al., 1978) Other reports also imply higher consunption of salty 
foods or table salt by blacks vs. whites (Karp et al., 1980; Kerr et 
al., 1982) . However, NHANES data do not show a greater use of table 



159 



salt among blacks compared to whites. Age-specific tabulations of 
salt use patterns of black and white adults tend to show lower levels 
among blacks than whites in most age groups (National Center for 
Health Statistics, 1985) . 

No differences in the proportions of calories from carbohydrate, 
protein, saturated and unsaturated fat between black and white males 
or females are apparent in the NHANES data for any age group bewteen 
1 and 74 (National Center for Health Statistics, 1985) . However, 
the NHANES data indicate a greater tendency towards meal skipping and 
less use of vitamin and mineral supplements among blacks than vrtiites 
(National Center for Health Statistics, 1985) . Regarding meal 
skipping, 20 to 23% of black males and females skip breakfast 
conpared with 12 to 14% of white males and females. Differences in 
percentages of blacks and whites skipping lunch, dinner, and snacks 
are of a similar order. Age-sex-race specific tabulations give the 
same general iirpression as the overall data. Frank et al. (1978) 
noted that fewer black than white boys and girls in the Bogalusa 
Study ate breakfast. Haider and Wheeler (1980) also noted meal 
skipping by black adolescent girls. NHANES data also indicate higher 
use of vitamin and mineral supplements among whites than blacks in 
the U.S. population (National Center for Health Statistics, 1985). 

Carittoean (West Indian) blacks have traditional diets similar in 
structure to those described for Hispanics. Puerto Rican diets are 
one type of Caribbean diet. The cultures and foods of various 
Caril±)ean islands include ^>anish, French, Chinese, Dutch, African, 
and Amerindian influences (James, 1978) . Examples of some Caribtean 
foods are shown in Table 12. 



Table 12: 

Examples of Some Caribbean Foods 



LEGUMES chick peas, cow peas, pigeon peas; peas; 
kidney fcieans, lima beans, soybeans 

VEGETABLES apio (root resembles carrot and parsnip); batata 

(starchy tuber) ; calabasa (punpkin and squash) ; chayote 
(resembles cucunt)er and squash) ; malanga (taro; also 
called dasheen, tania, tainer, tanyah, yautia malanga) ; 
pana (seedless breadfruit) ; pane de pepita (seed 
breadfruit) ; plantain (variety of banana, eaten baked, 
fried, or lx)iled) ; yautia (varieties include belembe, 
calalu, eddo) ; yuca (fleshy, tuberous, starchy root) 

FLAVORING AND basil, sweet chili pepper, sesame seeds, aniseed, 
SEASONING HERBS coriander, dill, ginger, sweet marjoram, 
spearmint, oregano, rosemary 



SOURCE: New York City Department of Health, 1976b 

160 



6.2 N utritiona l R isk 

6.2,1 Infancy and Childhood 

Data from the National Natality Surveys indicate a higher 
reported incidence of breast feeding in 1980 than in 1969 for infants 
born to married women in the United States (Fetterly and Graubard, 
1984) . Breast feeding incidence was higher in white women than black 
wraten respondeits in both years. Nine percent of black women vs. 19% 
of vrtiite women exclusively breast fed their infants in 1969; in 1980 
the percentages were 25% for black women and 51% for v^ite women. A 
higher incidence of breast feeding among women with higher levels of 
education was reported for white women in 1969 and 1980 and for black 
wranen in 1980. In 1969, the incidence of breast feeding was lower in 
black women with higher educational attainment. OSie racial 
differences are still seen after adjustment for education and 
parity. 

Similar findings of race differences which are not explained by 
education have been reported in studies among small sairples of black 
and white women (including some unirarried women) in different parts 
of the Uhited States (Rassin et al., 1984; Schaefer and Kumanyika, 
1985) . In these studies, the effect of maternal education on breast 
feeding is more prominent among whites than among blacks. A report 
of Russo et al. (1981) points out the need to distinguish native-bom 
from inmigrant blacks in this respect. Among several ethnic groups 
studied, breast feeding incidence was highest among West Indian 
blacks (64%) and lowest among native-bom blacks (15%) . Centers for 
Disease Control Pregnancy Nutrition Surveillance data for 1981 
indicated that, among low income postpartum women in their sanple for 
whom breast feeding data were available, breast feeding incidences 
were 16% and 37% for black women less than 20 years old and over 20, 
respectively, vs. 34 and 53% for vrtiite women in these two age 
categories (Centers for Disease Control, 1983) . 

Black and white mothers may differ in the types of solid foods 
they give their infants. Black infants in the Massactaossetts survey 
by Ferris et al. (1978) were more likely than white infants to be fed 
juices or soup, and less likely to be fed fruit or vegetables, with 
some differences in these patterns for infants less than three months 
vs. 3-6 months of age and according to social class. Black mothers 
surveyed in Philadelphia reported a greater frequency of high sodium 
feeding practices (earlier introduction of solid foods, adding salt 
to infant foods, and feeding relatively higher sodium foods as 
snacks) than vrtiite mothers. Some, but not all of these racial 
differences were statistically explained by black-white differences 
in matemal education (Schaefer and Kumanyika, 1985) . 



161 



An excess prevalence of linear grovrtii stunting is not evident in 
the Centers for Disease Control Pediatric Nutrition Surveillance 
System data (CDC-PNSS) for 1977-1982 for black children ages 2 years 
and older (although, as noted in section 2.3, inferences on this 
measure are confounded by differences in growth patterns for black 
and white children) (Centers for Disease Control, 1983; Trowbridge, 
1983) . In 1982, the percentage of black children below age 1 with 
low length for age was 9%, greater than the expectation of 5% below 
the 5th percentile and was higher than for whites, Hispanics and 
Native Americans. Prevalence of low length for age among black 
children in the 1 year age range was similar to that observed for 
black children less than one, but was lower than the prevalence among 
children in the other three ethnic groups and was lowest of the other 
three ethnic groups at ages 2,3, and 4 (Asian children were not 
reported in these data) (Trovrtjridge, 1983) . The same pattern was 
observed during 1977-1981, except that percaitages of Asian children 
less than two years old with low length for age were highest at all 
ages and for all years after 1977 conpared to the other ethnic groups 
(Centers for Disease Control, 1983) , 

Alvarez et al. (1984) reported a higher prevalence of short 
stature (ratio of height to NCHS median height) among black vs. 
Hispanic children under two years of age (27% vs. 10%) and a lower 
prevalence among black than among Hispanic children between ages 3 
and 12 years (<20% amcMig blacks and >25% among Hispanics) , The 
children studied were from a low income, inner-city health center 
population in Boston in 1978. Moderate undernutrition was reported 
for 16% of the black children in this sanple. 

A higher prevalence of low hemoglc±>in among black children ages 
two years and over is suggested in the CDC data for 1977-1981 
(Centers for Disease Control, 1983). Data from NHANES II indicate 
that mean hemoglobin levels for blacks are lower than for whites in 
all age-sex categories and in both poverty and non-poverty categories 
(National Center for Health Statistics, 1985) . However, 
differentials in the corresponding prevalences of anemia will depend 
on whether the same criteria are used for both blacks and whites (see 
section 2.3), Voors et al. (1981) reported that clinical aneonia, 
where observed in a sanple of children in Louisiana, occurred 
primarily in younger black boys and older black girls. Racial 
differences in blood levels for several nutrieits (e.g., higher mean 
serum copper, lower mean serum vitamin C, lower serum zinc (males)) , 
are evident in the NHANES II data (Naticaial Coiter for Health 
Statistics, 1985). However, the significance of these differences 
for the relative nutritional status of black and white children has 
yet to be fully interpreted. Tabulations of reported 24-hour dietary 
intakes for black and white children and adolescoits in NHANES II do 
not show large systematic differences in mean nutrient levels 
(National Center for Health Statistics, 1985) . 



162 



A 1981 survey of nutrient intakes among female adolescents in the 
South indicated higher consuiiption of several nutrients among white 
girls than among black girls (vitamins E, C, B-12, niacin, and 
folacin; calcium, phosphorous, magensium and zinc, expressed as 
individual totals and per 1000 kcal; and protein, vitamin D, and 
iron) (McCoy et al., 1984). Some of the racial difference was due to 
higher supplement use among white than among black girls. However, 
these data do not indicate that the diets of the black girls were 
inadequate in the nutrients listed. Most diets met the PDA's or were 
within the recoirinended ranges for the nutrients ascertained. 
Possible areas of excess risk among the black girls were intakes of 
folacin and vitamin D, Answers to a question on table salt use 
suggested higher salt use by black girls than white girls; however 
there were no racial differences in stated preferences for salty 
foods, 

Gartside et al, (1984) analyzed NHANES II dietary variables 
related to high density lipoprotein-cholesterol (HDL-C) levels in an 
atteitpt to idaitify determinants of the higher HDL-C levels in blacks 
conpared to whites. Variables considered were the Quetelet Index 
(QI: weight divided by the square of height), reported 24-hour intake 
of calories, protein, fat, carbohydrate, saturated fat, oleic acid, 
linoleic acid, cholesterol, and the ratio of linoleic to oleic acid 
(I/O) , expressed as totals and per kilogram of body weight. Using 
the data adjusted for body weight, there were no significant 
differences in intakes of any of the dietary cortponents studied for 
males ages 6 months through 20 years. Among females ages 6 months to 
5 years, total fat, oleic acid, linoleic acid, and the I/O ratio were 
significantly higher in blacks. Other statistically significant 
differences which were scattered across sex-age groups were not 
evident when intakes were expressed per unit weight. 

In the Gartside et al.(1984) analysis, the only significant 
racial difference observed in mean Quetelet Index (QI) estimates for 
children were higher QI values among black than among white girls in 
the 12 to 20 year age group. Among the low income children in the 
CDC-PNSS data for 1977-82, the prevalence of high weight for height 
among black children under age 2 was approximately twice the expected 
5% and was somewhat more prevalent than among white children in this 
data base. High weight for height was closer to the expected 
prevalence among preschool black children (2 to 5) and was somevrtiat 
higher than among white children in 1982 but not in 1977-1981 
(Centers for Disease Control, 1983; Trov^ridge, 1983). The 
prevalence of high weight for height among Hispanic and Native 
American children was higher than for black children at all ages 
after infancy. High weight for height was not in excess prevalence 
among the sairple of black children studied by Alvarez et al. (1984) . 



163 



6.2.2 Pregnancy 

Recent studies identifying excess nutritional risk specifically 
among black pregnant women were not identified. The extent to which 
possible problems in this area are being addressed may be available 
in WIC program data. Areas of concern for the nutritional status of 
black wcanen in general are discussed in section 6.2,3. 

Although not widely prevalent among black women in the general 
population, a noteworthy prevalence of pregnancy-associated-pica has 
been reported among some black women in the rural South and may still 
be a relevant concern. Vermeer and Frate (1979) reported that 57% of 
women, 28% of pregnant and postpartum women, and 16% of children of 
both sexes in a rural Mississippi county practiced geophagia, with an 
average daily clay consumption of 50 grams. Most of the children 
practicing geophagia were under age 4. CcMonunity norms included the 
use of clay as a "pacifier" for young children but suppression of the 
practice as children became older. Geophagia was not practiced by 
adolescents or adult men, but was coiniion amtaig reproductive-age and 
pregnant women. Other forms of pica were practiced by an additional 
19% of the people surveyed (primarily starch eating, but also eating 
of dry powdered milk) . The authors ' iirpression was that pica was not 
caused by or associated with dietary or nutritional deficiencies, 
although it may be indirectly related to appetite through an 
association with enotional well-being. 

Although various forms of pica are a hidden aspect of nutritional 
risk, Vermeer and Frate (1979) found little evidence of deleterious 
effects in their study. Hematocrit levels of women practicing 
geophagia were similar to those of non- practitiaiers . A possible 
aggravation of hypertension among pregnant women through excess 
sodium intake (either salt added to clay in preparation or consumed 
in baking soda) was noted. lUne authors also cite a 1975 case report 
(JAMA 1975; 234: 738) of possible pica-associated hyperkalemia among 5 
black patients in Washington D.C. with chronic renal failure. 



6.2.3 Adults 

Using either a "cut-off method" (12 mg/dl cutoff) or a "mixed- 
distribution" technique, Meyers et al. (1983) reported a higher 
prevalence of anania among black than white non-pregnant women in 
NHANES. Anenia prevalence estimates were low, 0.8 to 1.5 % of the 
total population for white women and 1.4 to 2.7% for black women; 
however, the authors noted that these percaitages represent a large 
number of U.S women. NHANES II data indicate lower mean hemoglobin 
and transferrin saturation levels among blacks than whites in all 
age-sex groups and lower levels of serum iron among black vs. white 
women. The serum biochaoistries also showed lower mean zinc levels in 
scane age groups of males and females, higher copper levels, and lower 
vitamin C levels for blacks than whites (National Caiter for Health 
Statistics, 1985) , 

164 



Windham et al. (1983a; 1983b) analyzed the nutrient density and 
nutriticaial adequacy of diets reported by respondaits in the National 
Food Consunption Survey, Multivariate adjustments for subgroup 
differences in distributions of age, sex, height, and weight were 
made to facilitate valid ccarparisons. Calcium density of reported 
diets was lowest among black and Hispanic respondents. Blacks had 
the lowest dietary magnesium density of any racial or ethnic group. 
Dietary vitamin A and thiamin densities were significantly higher for 
blacks than whites. Diets of blacks approximated (i.e. were within 
80% of) the nutrient density standard for all nutrioits studied. 

Inspection of NHANES II tabulations (National Center for Health 
Statistics, 1985) of nutrient intakes among black and white adults 
suggests differences in mean intakes of several nutrients, but often 
at levels which approximate dietary standards, i^parent lower 
intakes of calcium and phosphorous, potassium, and iron among blacks 
in many of the adult sex-age groups may be inportant. However, 
statistical tests of the significance of these differences have not 
been published. The Gartside et al. (1984) NHANES II analysis of 
possible determinants of black^white HDL-C differences indicated 
significantly higher calorie, protein, and fat intakes per kilogram 
of body weight in white vs. black men ages 21 to 65 and among white 
vs. black wcanen ages 25 and over. Total caloric intakes were also 
higher among white vs. black women, in contrast to the significantly 
higher Quetelet Index (QI) of black women at all ages over 21. 
Significant differences in QI are not seen among black and white men. 

The Gartside et al. findings regarding higher relative weight 
among black vs. white women are indicative of what is clearly the 
most striking nutrition-related disparity between blacks and 
vrtiites — the marked excess prevalence of obesity among black women vs. 
white womai and vs. white and black men. A graphic presentation of 
the extent of the problem is shown in Figure 1. Using the iiiplied 
criterion of a 15% expected prevalence high relative weight, the 
excess prevalence among black women is approximately 25% vs. an 
excess of only 10% among white women and men. The corresponding 
overall prevalences are approximately 40% among black women and 25% 
among tte other race-sex groups. The analyses of Gartside et al. 
(1984) do not support the hypothesis that this excess prevalence of 
overweight among black women is caused by higher calorie or fat 
consunption among black vs. white women, either in terms of total 
intakes or per unit of body weight. In fact, the opposite appears to 
be true. 



165 




to 



a> 
o 



^ 



■^ 

01 

3 

> 

o 



o 

•o- 

I 

o 

CM- 

in 
01 
en 
re 

s 

X! 

re 










^^^^ 








J$$^$$$$^§S$^ 














^ 


V//. 


'/A 


/ 



0) 


C 




1- 


0) 




= 


o 




en 


s_ 




<- 


QJ 






~r 

o 






in 




I I I I I J 



I I r I I I 



O 



o 
m 



o 
ru 











C 










c 










•^ 










•bJ 










re 










c 










-,- c 










tn E O 










E ro •.- 




r 






re • X 4-) 






CO 


s_ oo UJ -t- 




i 




OJ 

E 


o =) j: jj 
— X) — s 








^ .— fO 




u. 


o 


0) 


O QJ -O 

c 1 :e: c 




2 


m 
1 




ro 1— 






(D 


-C 


+J QJ fO -C 








S 


^ >5 C ^ 






II 


cn o ' — 






s: 


•1- CTi ■•- fD 




i 


^H 


3 


0) CM 4-J OJ 




m 






3 1 TO 3: 

II o z: 












3 CM •- 










t/) 


QJ (O 










<u 


QJ QJ j: c 
1- J= -^ o 




u. 






ra 

E 
QJ 
4- 


OJ *J -1- 




m 






-c E -1-^ 








^ 


S CT) O (0 








C S- 2 








n. -f- u- 










^ I/) QJ 








<U 


^ = QJ ^ 








+-> 


3 s- 4-= 






,^p_ 


•1 — 


•* re 




z 

2 






-C 


^ c: E 








2 


O QJ C\J O 








II 


E ^o s- 










U- 


QJ O 1 4- 










3 


-- 3 O 










■tJ S_ Ol S- 




L. 


■T 




QJ '^- 




i 


1 




O S- O 

s- Ln o CO 






^-i 




QJ • M- 1 






r^ 


Ol 


C— 1 VO 






01 


1 — 


ro r^ 




Z 


(O 


.C X) J-" CTi 




^^ 


E 


-tJ c tc — 




UJ 






U-> ro O 


tn 








-^ 


CO -o — 


U 








u 


C C — . 


•*— 








fO 


OJ QJ ■ fO 


4-" 










J= E CL T3 


in 








Q 


•^ =! "3- C 


•*- 


u. 






II 


i- O t^ Its 


■^ 


m 






1/1 O S- 1 


ro 






•^ U- Ol — < — 










-u CM QJ cr> in 










x; 11 u -^ QJ 


-C 








en Q. c f— 


4-J 






LO 


— QJ S- O 


^-^ 






^ 


OJ 


a' -ei s_ o >, 


ro 


T 






1 


3 C QJ H- (_> 


CU 


2 






IT3 


S_ ro »*- 


ii 






E 


QJ QJ fC - 










OJ 


> ♦■ i- J-J t/1 


i_ 








M- 


o i^ ro >, 

i- QJ Q QJ 


o 








1^ QJ X > 




Ll 


rvi 


o 


O -W -^ i_ 


i- 


2 


to 


ro 


QJ • n 
c E 1/1 • — -oo 


QJ 




o 


-Q 


o ro oo 

•1- C UJ c 


OJ 




ID 


II 

Ll_ 


s- •>- c =: o 


(_J 




01 


QJ o z — 




m 


^-t 


CQ 


.IJ 4-> .,- *J 


^^ 






— jr ^ fD 


ro 






s- en ro >, E 


c 










U -t- t— QJ -1- 


o 








Q 


oj = > E 










Z 


QJ -c El s_ ro 


4-> 








LlJ 


j: II o => X 


ro 


Ll 

m 






CJ3 


1— = C t/1 UJ 


z 






UJ 
1 












UJ 

1— 
O 

2 


UJ 












to 



uoL^PL^dod :^L"P^ J-O au30J3d 



If 



6.3 Conclusions 

The overall structural similarities between the traditional diets 
of black and white Americans lead to an expectation of more 
similarities than differences in nutritional status. The literature 
and tables reviewed are generally in line with this expectation. 
Intakes of food in the black population may be somewhat less than 
among whites leading to lower overall intakes of many nutriaits. 
However, a distinction should be made between intakes that are lower 
than those of whites vs. those that are lower than standards of 
dietary adequacy. 

The dietary and nutritional status data from NHANES II have been 
published but not yet fully interpreted. No major black-white 
disparities on these measures could be defined, although concern for 
the adequacy of iron, calcium, magnesium, and potassium intakes among 
segments of the black population is suggested by the national data. 
A discussion of issues related to the interpretation of dietary 
calcium adequacy is included in section 2.3. Inadequate iron intakes 
among black women are reflected in their excess prevalence of anemia, 
even after the appropriate adjustments for differences in hemoglobin 
distributions of blacks and whites. To the extent that nutritional 
adequacy among whites is a function of the use of vitamin 
suEpleroents, blacks — ^who are less likely than whites to use 
suEplenents — will be at higher nutritional risk. 

Within the limitaticMis of interview data for determination of 
sodium intake, the NHANES data do not indicate that sodium intakes of 
blacks are higher than those of whites. However, small studies 
indicate that a greater preference for and intake of certain highly 
salted foods among blacks is prc^aable. The quantitative contribution 
of these foods to overall sodium intakes in blacks is unclear. 

Breast feeding incidence has increased among blacks but is still 
much lower than among whites. Depending on the type of milk 
substitutes used and acconpanying patterns of feeding solid foods, 
this lower incidence of breast feeding may put black infants at a 
relative disadvantage vs. white infants. The nutritional status of 
reproductive-age black women is of continuing interest relative to 
the excess prevalence of low birth weight among blacks. lUnere are 
potential nutritional risks to both mother and fetus associated with 
the practice of pica among black pregnant women. Pica may also be a 
concern for some subgroups of black children. 

A clear picture of nutritional risk among black children was not 
obtained. Neither linear growth stunting nor obesity was in excess 
among low-income black children (over age 1 or 2) in the Centers for 
Disease Control Pediatric Nutrition Surveillance data base. Most of 
the recent smaller studies identified were amcaig low-income children 
and could not support generalities to the population of black 
children as a whole. Excess sodium intake among black girls is 
implied. Meal skipping by both boys and girls may be problenatic. 

167 



The most outstanding nutritional problem in the black adult 
population is the excess prevalence of ctoesity among black women, a 
trend which is not necessarily evideit in younger black girls but 
which appears during adolescence. Data on reported dietary intakes 
do not explain this excess <±»esity. Further study of meal patterns 
and activity levels as well as aiergy metabolism among black women 
are indicated. Meal skiH>ing, which appears to be in excess among 
blacks of both sexes and at all ages, may coitribute to the obesity 
in black women. 



168 



7. SUMMABY 

The traditional dietary patterns of Asian, Hispanic, and Native 
Americans are structurally differoit from diets characteristic of 
white Americans in that carbohydrates rather than meats are the major 
sources of kilocalories. The diets of these minority subgroups also 
include many unique flavoring principles, food varieties, and food 
preparation methods. Diets of black Americans are gaierally similar 
to those of white Americans, particularly in the southeastern U.S. 
Specifically-black dietary elements relate to foods and cooking 
methods used by blacks during slavery. 

Nutritional risks associated with minority group dietary patterns 
are gaierally not inherent in traditional food patterns (except for 
the high dietary sodium contents of scxne of the traditional foods) . 
Potential risks are determined by social, economic, or acculturation 
factors (see Table 3) . Food ideologies which may influence 
health- related food behaviors are seen in all cultures, e.g., the 
"hot-cold" dichotomies of Asian and Hispanic cultures. Nutritional 
risks which may derive frcMn strict adherence to certain food 
ideologies include unfavorable food restrictions among pregnant wcanen 
or young children. Although it is doubtful that such food beliefs or 
associated restrictions persist at levels which have public health 
significance, certain subgroups may be vulnerable in this respect. 
Similarly, the traditional practice of consuming non-food substances 
(pica) among pregnant wcxnen and young children may be a concern in 
certain minority subgroups. 

The available data indicate that nutritional risk in U.S. 
minority populations is related to the following factors: tte 
income-related limitations on the quantity and quality of food 
consumed, the excess prevalences of obesity and obesity-related 
diseases in scane minority subgroups, and the shifts toward dietary 
patterns which have been epidaniologically associated with high rates 
of cancer and cardiovascular disease. In addition to these general 
areas of risk, some specific nutritional concerns evident in the data 
for each minority group have been identified (see sections 3.3, 4.3, 
5.3 and 6.3) . With the caveat that the problems identified are a 
function of the data available, these group-specific nutritional 
problon areas can be sumnarized as follows: 

- among Asian Americans, evident areas of concern are the ccarpromised 
growth and overall nutritional status of iinnigrant subgrcHips; and 
to carryovers of the high sodium intakes from traditional Asian 
diets. 

- among American Indians, evident concerns relate to child growth, to 
excess alcohol intake, cAsesity and diabetes; 

- among Hispanic Americans, evident concerns are growth stunting, 
obesity, infant feeding patterns, and adequacy of dietary vitamin 
A; 

169 



- among black Americans, evident concerns relate to vitamin and 
mineral intakes (vitamin A, vitamin C, iron, calcium, magnesium, 
potassium) , to high sodium intakes, to infant feeding 
patterns, and to the excess prevalence of obesity among black 
women. 



In addition to the numerous methodological issues in nutritional 
assessment, there are several cross-cutting issues which ^^ly 
specifically to the validity of nutritional risk inferences from data 
on non-vdiite populations. These issues, as outlined in section 2.3, 
are of three general types: 1) issues related to the applicability 
of standards and references used to interpret dietary, 
anthropometric, and biochemical measures of nutritional status; 2) 
possible differences in sensitivity to a given level of nutriait 
excess between whites and non-whites due either to heredity factors 
or to differences in levels of nutrition-related morbidity; and 3) 
the large socioeccaiomic status effects within minority group data. 

Although nutrition, as such, is not a separate area of aiphasis 
by a Task Force subgroup, considerations discussed in this review 
have iirplications that may be relevant to the deliberations of 
several Task Force subcommittees. The iirportance of standardized 
data on dietary and nutritional status measures for sufficiently 
large and representative sairples of each minority subgroup is clear. 
There is also a need to ensure that dietary and nutritional status 
reference data are applicable multiculturally. Many nutritional 
effects of potential public health iirportance are related to subtle 
variations within characteristic U.S. food intake patterns. 
Monitoring race-related nutritional risks will require approaches 
which are sensitive to the often subtle differences between nutriait 
intakes of minority group Americans and whites. 



170 



LITERATURE CITED 

Acosta, PB; Aranda, RG; Lewis, JS; and Read, M; (1974) Nutritional 
status of Mexican-American preschool children in a border town. 
Am J Clin Nutr 27(12) : 1359-1368 

Adrian, J; Daniel, R; (1976) Impact of socioeconomic factors on 
consunption of selected food nutrients in the United States. 
Am J Agr E Con 58:31-38 

Alford, BB; Nance, EB; (1976) Customary foods in the Navajo diet. 
J Am Diet Assn 69(5) :538-539 

Alvarez, SR; Herzog, LW; Dietz, W Jr; (1984) Nutritional status of 
poor black and Hispanic children in an urban neighborhood health 
Center. Nutr Res 4:583-589 

American Heart Association Nutrition Committee; (1982) Rationale of 
the Diet-Heart Statement of the American Heart Association. 
Circulation 65:839A-854A 

Asian/Pacific Islander Task Force on High Blood Pressure Education 
and Control; (1984) Social and Cultural Considerations in the 

Treatment of Hypertensives in Selected Asian/Pacific Islander 
Populations. 310 8th Stret, Suite 305B. Oakland, CA 94607. 

Avioli, LV; (1980) Major minerals. A. Calcium and phosphorus. In: 
Goodhart, RS; Shils, ME; eds. Modern Nutrition in Health and 
Disease . Lea and Febiger, Philadelphia, PA. Chapter 7, pp 
294-309 

Bass, MA; Wakefield, M; (1974) Nutrient intake and food patterns of 
Indians on Standing Rock Reservation. J Am Diet Assn 64:36-41 

Bowering, J; Lowenberg, RL; Morrison, MA; Parker, SL; Tirado, N; 

(1978) Infant feeding practices in East Harlem. J Am Diet Assn 
72(2): 148-154 

Bryant, CA (1982) Impact of kin, friend, and neighbor networks on 
infant feeding practices. Soc Sci Med 16:1757-1765 

Burroughs, AL; Huenemann, RL; (1970) Iron deficiency in rural infants 
and children. J Am Diet Assn 57:122-128 

Butte, NF; Calloway, DH; and VanDuzen, JL; (1981) Nutritional 

assessment of pregnant and lactating Navajo women. Am J Clin 
Nutr 34:2216-2228 

California Department of Health Services. Maternal and Child Health 
Branch. 714/744 P Street, Room 300, Sacramento, CA 95814. (1975) 
"Ethnic, Social and Economic Influences on Diet Patterns". In: 
Nutrition During Pregnancy and Lactation. Chapter 8, pp. 73-93. 



171 



Cardenas, J; Gittos, CE; Young, EA; (1976) lAitritional beliefs and 
practices in primigravid Mexican-American wcanen. J Am Diet Assn 
69(3):262-265 

Casey, P; Harrill, I; (1977) Nutrient intake of Vietnamese women 
relocated in Colorado. Nutr Rep Int 16:687-693. 

Centers for Disease Control (1983) Nutrition Surveillance. 1981 . 

U.S. Department of Health and Human Services. HHS Pub. No. (CDC) 8 

Chase-the-Bear, R; Bonnell, M; Morse, HG; Rate, RG; (1979) Hopis and 
Navajos not lean. N Eng J Med 13:301(24) :1348 (letter) 

Chew, T; (1983) Sodium values of Chinese condiments and their use in 
sodium-restricted diets. J Am Diet Assn 82(4) : 397-401 

Clifford, NJ; Kelly, JJ; Leto, TF; Eder, HA; (1963) . Coronary heart 
disease and hypertension in the White Mountain i^che Tribe. 
Circ 8:926-931. 

Committee on Diet, I*itrition and Cancer (1982) Diet, Nutrition and 
Cancer. National Research Council, National Academy of Sciences. 
Washington, D.C. 

Cook, CB; Eiler, DA; and Forker, (X); (1975) Beverage consumption 
patterns in New York State. J Am Diet Assn 67:222-227 

Comely, PB; Bigman, SK; Watts, TX>; (1963) Nutritional beliefs among 
a low-inccane, urban population. J Am Diet Assn 42:131-135 

Crane, NT; Green, NR; (1980) Food habits and food preferences of 
Vietnamese refugees living in northern Florida. J Am Diet Assn 
76(6):591-593 

Cronin, FJ; Krebs-Smith, £M; wyse, BW; Light, L; (1982) 

Characterizing food usage by demographic variables. J Am Diet 
Assn 81:661-673 

DeStefano, F; Coulehan, JL; Wiant, MK; (1979) Blood pressure survey 
on the Navajo Indian reservation. Am J Epi 109:335-345 

Dewey, KG; Chavez, MN; Gauthier, CL; Jones, LB; and Ramirez, RE; 
(1983) Anthropometry of Mexican-American migrant children in 
northern California. Am J Clin Nutr 37(5): 828-33 

Dietary Guidelines for Americans (1980) U.S. Depts. of Agriculture 
and Health and Human Services. Home and Garden Bulletin No 232. 

Doeblin, TD; Evans, K; Ingall, GB; Dowling, K; Chilcote, ME; Elsea, 
W; Bannerman, RW; (1969) . Diabetes and hyperglycemia in Seneca 
Indians. Hum Hered 19:613-627. 



172 



Enterline, J; (1984) Personal communication regarding preliminary 
results of a case-control study of prostate cancer among black 
men. Howard Lftiiversity. Washington, D.C. 

Farmer, ME; White, LR; Brody, JA; Bailey, KR; (1984) Race and sex 
differences in hip fracture incidence. Am J Pub Health 
74:1374-1380 

Ferris, AG; Vilhjalmsdottir, LB; Beal, VA; and Pellett, PL; (1978b) 
Diets in the first six months of infants in western 
Massachusetts. II. Semi-solid foods. J Am Dietet Assn 
72:160-163. 

Fetterly, K; Graubard, MS; (1984) Racial and Educational Factors 
Associated with Breast Feeding - United States, 1969 and 1980. 
MMWR 33(11): 153-154 

Food and Nutrition Board (1980) Toward Healthful Diets. National 
Research Council, National Academy of Sciences. Washington, D.C. 

Forman, MR; Graubard, BI; Hoffman, HJ; Beren, R; Harley, EE; Bennett, 
P; (1984) The Pima Infant Feeding Stuc^: Breast feeding and 
gastroenteritis in the first year of life. Am J Epi 119(3): 
335-349 

Frank, GC; Berenson, GS; and Webber, LS; (1978) Dietary studies and 
the relationship of diet to cardiovascular disease risk factor 
variables in 10 year old children - The Bogalusa Heart Study. Am 
J Clin Nutr 31:328-340 

Frerichs, RR; Webber, LS; Srinivasan, SR; Berenson, GS; (1977) 
Hemoglobin levels in children from a biracial southern 
community. Am J Pub Health 67(9) : 841-845 

Frishancho, AR; Leonard, WR; and Bollettino, LA; (1984) Blood 
Pressure in Blacks and Whites and its relationship to dietary 
sodium and potassium intake. J Chron Dis 37:7; 515-519 

Fulmer, HS; Roberts, BW; (1963) Coronary heart disease among the 
Navajo Indians. Ann Int Med 59:740-764 

Galli, N (1975) The influence of cultural heritage on the health 
status of Puerto Ricans. J Sch Health: 45(1): 10-16 

Garcia-Palmeri, MR; (1973) Precursors of coronary artery disease in 
Puerto Rico. Am J Clin Nutr 26(10) : 1133-1137 

Garcia-Palmeri, MR; Tillotson, J; Cordero, E; Costas, R; Sorlie, P; 

Gordon, T; Kannel, WB; Colon, AA; (1977) Nutrient intake and 

serum lipids in urban and rural Puerto Rican men. Am J Clin Nutr 
30(12): 2092-2100 



173 



Gam, SM; Clark, DC; (1976) Problems in the nutritional assessment of 
black individuals. Am J Pub Health 66:262-267 

Gartside, PS; Khoury, P; and Glueck, CJ; (1984) Determinants of HDL 
cholesterol in blacks and Whites: the second national HANES. Am 
Heart J 108: #3 (2): 641-653 

Gillum, RF; (1979) Pathophysiology of hypertension in blacks and 
whites. A review f the basis for racial blood pressure 
differences. Hypertension 1:468-475 

Gillum, RF; Prineas, RJ; Palta, Ma; Horibe, H; (1980) Blood pressures 
of urban Native American school children. Hypertension 2:744-749 

Gillum, RF; Gillum, BS; Smith, N; (1984) Cardiovascular risk factors 
among urban American Indians. Blood pressure, serum lipids, 
smoking, diabetes, health knowledge, and behavior. Am Heart J 
107:756-776 

Gite, L; Perry, J; (1983) Diet Rightl Losing weight soulfully. 
Essence Magazine 14(4) : 117-118 

Gonzalez, NL; (1972) "Changing dietary patterns of North American 
Indians". In: Moore WM et al.; eds. Nutrition, Growth, and 
Developnent of North American Indian Children. HEW Pub. NO. 
(NIH) 72-26. Washington, D.C. 

Gordon, AM Jr.; (1982) Nutritional status of Cuban refugees: a field 
study on the health and nutriture of refugees processed at Opa 
Locka, Florida. Am J Clin Nutr 35(3) :582-590 

Graham, S; Haughley, B; Marshall, J; Priore, R; Byers, T; Rzepka, T; 
Mettlin, C; Pontes, JE. (1983) Diet in the epidemiology of 
carcinoma of the prostate gland. J Natl Cancer Inst 70:687-692. 

Grivetti, LE; Paquette, MB; (1978) Nbntraditional ethnic food choices 
among first generation Chinese in California. J Nutr Ed 
10(3): 109-112 

Haider, SQ; Wheeler, SH; (1980) Dietary intake of black and Hispanic 
teenage girls. J Am Diet Assn 77(6) :677-681 

Hanes, V;Vermeersch, J; Gale, S; (1984) The National Evaluation of 
School Nutrition Programs. Program inpact on dietary measures. 
Am J Clin Nutr 40 (supp) : 390-413. 

Hanis, CL; Ferrell, RE; Barton, SA; Aquilar, L; Garzaibarra, A; 

Tulloch, BR; Garcia, CA; and Schull, WJ; (1983) Diabetes anong 
Mexican-Americans in Starr County, Texas. Am J Epi 118:5:659-672 

Hankin, JH; Nomura, A; and Rhoads, GG; (1975) Dietary patterns among 
men of Japanese ancestry in Hawaii. Cancer Res 35:3259-3264 



174 



Homer, MR; Olson, C; and Pr ingle, DJ; (1977) Nutritional status of 
Chippewa Head Start children in Wisconsin, Am J Pub Health 
67(2): 185-186 

Hunt, IF; Murphy, NJ; Gomez, J; and Smith, JC Jr.; (1979) Dietary 
zinc intake of low- income pregnant women of Mexican descent. Am 
J Clin Nutr 32(7) : 1511-1518 

James, SM; (1978) When your patient is black West Indian. Am J 
Nursing 78(11) : 1908-1909. 

Jercane, NW; Pelto, GH; (1981) Integrating ethnogr^hic research with 
nutrition studies. Fed Proc 40:2601-2605 

Johnston, FE; McKigney, JI; Hopwood, S; and Smelker, J; (1978) 
Physical growth and developtnent of urban Native Americans; a 
stu<^ in urbanization and its iiiplications for nutritional 
status. Am J Clin Nutr 31(6) : 1017-1027 

Judkins, RA; (1978) American Indian medicine and conterrporary health 
problems. IV. Diabetes and perception of diabetes among Seneca 
Indians. NY State J Med 78:1320-1323 

Karp, RL; Williams, C; and Grant, JO; (1980) Increased utilization of 
salty food with age among preteenage black girls. J Natl Med 
Assn 72:3;197-200 

Kerr, GR; Amante, P; Decker, M; and Callen, PW; (1982) Ethnic 
patterns of salt purchase in Houston, Texas. Am J Epi 
115:906-916 

Kim, KK; Kohrs, MB; Twork, R; Grier, MR; (1984) Dietary calcium 
intakes of elderly Korean Americans. J Am Diet Assn 
84(2): 164-169 

Knishbacher, S; Porter, DV ; (1983) Nutrition and Diet-Related 

Diseases. Congressional Research Service. Library of Congress. 
Washington, D.C. (1981 document revised and updated by Porter in 
1983) 

Koh, ET; (1980a) Selected blood ccaiponents and urinary B vitamins as 
related to age and sex of the black population in southwest 
Mississippi. Am J Clin Nutr 33:670-676 

Koh, ET; Chi, MS; and Lowenstein, EW; (1980b) Conparison of selected 
blood components by race, sex, and age. Am J Clin I^tr 
33:1828-1835 

Kolonel, LN; Nomura, AMY; Hirohata, T; Hankin, JH; and Hinds, MW; 

(1981) Association of diet and place of birth with stcanach cancer 
incidence in Hawaii Japanese and Caucasians. Am J Clin Nutr 
34:2478-2485 



175 



Kramer, HJ; Hunter, R; Rosse, C; Shiith, NJ; (1977) Race related 
differences in peripheral blood and in bone marrow cell 
populations of American black and American white infants, J Natl 
Med Assn 69:327-331 

Kreidler, PL; Boehm, WT; Lentner, MN; and Driskell, JA; (1980) Dairy 
product purchase by southern households. J Am Diet Assn 77:41-46 

Kuhnlein, HV; Calloway, DH; and Harland, BF; (1979) Composition of 
traditional Hopi foods. J Am Diet Assn 75(1): 37-41 

Lackey, CJ; (1983) Pica during pregnancy. Conteiporary Nutrition. 8 
(11) 

Larson, LB; Dodds, JM; Massoth, DM; and Chase, HP; (1974) Nutritional 
status of children of Mexican-American migrant families. J Am 
Diet Assn 64:29-39 

Lewis, JS; Glaspy, MF; (1975) Pood habits and nutrient intakes of 
Filipino women in Los Angeles. J Am Dietet Assn 67:122-125. 

Lieberman, LS (1979) Medico-nutritional practices among Puerto-Ricans 
in a small urban north-eastern community in the US. Soc Sci Med: 
13B: 191-198 

Lowenstein, FW; (1981) Review of nutritional status of Spanish 

Americans based on published and unpublished reports between 1968 
and 1978. World Rev Nutr Diet 37:1-37 

Ludman, EK; Newman, JM; (1984) Yin and Yang in the health-related 
food practices of three Chinese groups. J Nutr Ed 16(1): 3-5 

Magnus, PD (1983) Breastfeeding among Hispanics. Am J Public Health 
73(5) :597 (letter) 

Marrs. DC; (1978) Milk drinking by the elderly of three races. J Am 
Diet Assn 72(5) : 495-498 

Ma^Derry, RH; Lindeman, RD; (1963) A survey of chronic disease and 
diet in Seninole Indians in OklahcHia. Am J Clin I*itr 13:127-134 

McCoy, H; Kenney, MA; Kirby, A; Disney, G; Ercanli, FG; Glover, E; 
Korslund, M; Lewis, H; Li^^man, M; Livant, E; Moak, S; Stallings, 
S; Wakefield, T; Schilling, P; Ritchey, SJ; (1984) Nutrient 
intakes of female adolescents from eight southern states. J Am 
Dietet Assn 84:1453-1460 

McGee, D; Rhoads, G; Hankin, J; Yano, K; and Tillotson, J; (1982) 
Within-person variability of nutrient intake in a group of 
Hawaiian men of Japanese ancestry. Am J Clin Nutr 36(4) : 657-663 



176 



Meyers, LD; Habicht, JP; Johnson, CL; (1979) Cornponents of the 

differencein hemoglobin concentration in blood between black and 
white women in the ttiited States. Am J Epi 109(5) : 539-549 

Meyers, I£); Habicht, JP; Johnson, CL; Brownie C; (1983) Prevalences 
of anemia and iron deficiency anemia in black and white women in 
the liiited States estimated by two methods. Am J Pub Health 
73:1042-1049 

National Center for Health Statistics (1982) Diet and dental health, 
A study of relationships. United States, 1971-1974. Vital and 
Health Statistics, series 11. Data from the National Health 
Survey (No, 225) (DHHS Pub (PHS) 82-1675) . 

National Center for Health Statistics; (1983a) Health Lhited States ^ 
1983. Tables 30 (pp 129-130) and 31 (pp 131-132) . DHHS Pub. No. 
(PHS) 84-1232. Public Health Service. Washington. U.S. 
Government Printing Office, Dec. 1983 

National Center for Health Statistics. S. Abraham: 1983b; Ctoese and 
overweight adults in the Uiited States. Vital and Health 
Statistics. Series 11, No. 230. DHHS Pub. No. 83-1680. Public 
Health Service. Washington. U.S. Goveminent Printing Office. 

National Center for Health Statistics (1985) Caloric, selected 
nutrient, and food intake of persons 1-74 years of age, by race 
and sex. United States 1971-1974 and 1976-1980. Vital and 
Health Statistics, series 11 (unpublished draft) 

National Institutes of Health; (1984) Osteoporosis. Consensus 
Development Conference Statement. Vol 5, #3; US Government 
Printing Office, Washington, D.C. Stock #1984-421-132:4652 

National Research Council (1980) Recommended Dietary Allowances . 
Ninth Edition, National Academy of Sciences, Washington, D.C. 

Neaton, JD; Kuller, LH; Wentworth, D; Borhani, NO; (1984) Total and 
cardiovascular mortality in relation to cigarette smoking, serum 
cholesterol concentration, and diastolic blood pressure among 
black and white males followed up for five years. Am Heart J 
108:759-770 

New York City Department of Health. Bureau of Nutrition Services. 
93 Worth Street, Room 714, New York, NY 10013; (1981) 
Characteristics of the Puerto Rican Food Habits 

New York City Department of Health. Bureau of Nutrition Services. 
93 Worth Street, Room 714, New York, NY 10013; (1976a) 
Description of Puerto Rican Foods. 

New York City Department of Health. Bureau of Nutrition Services. 
93 Worth Street, Room 714, New York, NY 10013; (1976b) Glossary 
of Caribbean Foods. 

177 



Newcomer, AD; Thomas, PJ; McGill, DB; Hofmann, AF; (1977) Lactase 
deficiency: a ccanmon genetic trait of the American Indian. 
Gastroenterology 72 (2) : 234-237 

Nguyen, TIj Do, TH; Craig, WJ; Ziinnerman, G; (1983) Food habits and 
preferences of Vietnamese children. J School Health 
53(2): 144-147 

NOmura, A; Henderson, BE; and Lee, J; (1978) Breast cancer and diet 
anong the Japanese in Hawaii. Am J Clin Nutr 31(11): 2020-2025 

Owen, GM; Yanochik-Owen, A; (1977) Should there be a different 
definition of anemia in black and vrtiite children? Am J Pub 
Health 67:865-866 

Owen, GM; Garry, PJ; Seymoure, RD; Harrison, GG; Acosta, PB; (1981) 
Nutrition studies with White Mountain i^che preschool children 
in 1976 and 1969. Am J Clin Nutr 34(2) :266-277 

Paige, DM; Bayless, TM; Ferry, (3D; Graham, GG; (1971) Lactose 
malabsorbtion and milk rejection in Negro children. Johns 
Hoj^ins Med J 119:163-169 

Parker, SL; Bowering, J; (1976) Folacin in diets of Puerto Rican and 
black women in relation to food practices. J Nutr Ed 8(2), cited 
in Sanjur (1982) , p. 246 

Pettitt, DJ; Lisse, JR; Knowler, WC; and Bennett, PH; (1982) 

Mortality as a function of c±>esity and diabetes mellitus. Am J 
Epi 115(3): 359-366 

Rassin, DK; Richardson, CH; Baranowski, T; Nader, PR; Guenther, N; 
Bee, DE; Brown, JP; (1984) Incidence of breast feeding in a low 
socioeconomic group of mothers in the United States. Ethnic 
patterns. Pediatrics 73:132-137 

Report of the President's Task Force on Food Assistance (1984) 
Washington, D.C. 

Rhoades, SJ; (1982) St Regis-Mohawk Health Services Needs Assessment 
of Nutrition Services, pg. 5 (New York State Public Health 
Nutritionists report included in briefing book for the New York 
State Nutrition Watch Commission 

Roe, DA; (1973) A Placrue of Com. The Social History of Pellagra . 
Ithaca, New York. Cornell University Press 

Russo, Ml; Patel, R; Laude, TA; Rajkumar, SV; Gururaj, VJ; (1981) 

Infant feeding practices by ethno-cultural grouping. J Med Soc NJ 
78:737-740. 



178 



Salathe, LE; Gallo, AE; Boehm, WT; (1979) The inpact of race on 
consumer food purchases. USDA ESCS-68 Washington, DC 

Sanjur, D (1982) Social and Cultural Perspectives in Nutrition . 
Prentice-Hall. Inc. NJ. 

Schaefer, U; Kumanyika, SK; (1985) Maternal variables related to 
potentially high sodium infant feeding practices. J Am Dietet 
Assn 85:433-438. 

Seger, MT; Gil±)S, CE; Young, EA; (1979) Attitudes about breastfeeding 
in a group of Mexican-American primigravidas . Texas Med 
75(1): 78-80 

Shutte, JE; (1980) Growth standards for blacks. Current status. J 
Natl Med Assn 72:973-978 

Sievers, ML; (1968) Cigarette smoking and alcohol usage by 
southwestern American Indians. Am J Pub Health 58:77-82 

Sievers, ML; Fisher, JR; (1981) Diseases of North Americans Indians. 
In: Rothchild, HR; ed. Biocultural Aspects of Disease . Academic 
Press, Inc. Chapter 8 pp 191-213 

Siirpson, M; (1982) Annual Report of Pilot Nutrition Programs to 

Onondaga Indian Nation, pp 1-2. Cited in briefing materials for 
New York State Nutrition Watch Comnittee, 1982) 

Slonim, A; Kolassa, KM; and Bass, MA; (1981) The cultural 

appropriateness of the WIC program in Cherokee, NC. J Am Diet 
Assn 79:164-168 

Smith, JC; Mhango, CG; Warren, CW; Rochat, IW; Huffman, SL; (1982) 
Trends in the incidence of breastfeeding for Hispanics of Mexican 
origin and Anglos on the US - Mexico border. Am J Pub Hlth 
72(1):59-61 

Show, LF; (1976) "High Blood" is not high blood pressure. Urban 
Health 6:54-55 

Stem, MP; Pugh, JA; Gaskill, SP; and Hazuda, HP; (1982) Knowledge, 
attitudes, and behavior related to obesity and dieting in Mexican 
Americans and Anglos: the San Antonio Heart Study. Am J Epi 
115(6): 917-928 

Stem, MP; Gaskell, SP; Hazuda, HP; Gardner, LI; and Haffner, SM; 
(1983) Does obesity explain excess prevalence of diabetes among 
Mexican-Americans? Results of the San Antonio Heart Study. 
Diabetologia 24:272-277 

Stephenson, LS; Latham, MC; Jones, DV; (1977) Milk consuirption by 
black and white pupils in two primary schools. J Am Diet Assn 
71:258-262 

179 



Stoner, MK; Grivetti, LE; (1978) Vietnamese refugee. Mjustroait, 
diet and food behavior. Read before the Society for i^lied 
Anthropology, Mexico. 

Striinbu, JL; Sims, OS Jr.; (1974) A university system drug profile. 
Int J Addiction 9:569-583 

Suitor, IiW; Crowley, MF; (1984) "Promoting sound eating habits in 
different sociocultural situations". In: Nutrition. Principles 
and Applications in Health Programs. Second Edition. J.B. 
Lippincott, Philadelphia, PA Chapter 9, pp 91-105 

Tillotson, JL; Kato, H; Nichaman, MZ; Miller, DC; Gay, M; (1973) 

Heart disease and stroke in Japanese men living in Japan, Hawciii, 
and California: Methodology for conparison of diet. Am J Clin 
Nutr 26:177-184 

Tcxna, RB; Curry, ML; (1980) North Dakota Indians' traditional foods. 
J Am Diet Assn 76(6) :589-590 

Trowbridge, PL; (1983) Prevalence of growth stunting and obesity: 
Pediatric Nutrition Surveillance System, 1982. MMWR 32(4SS): 
23SS-26SS 

U.S. Bureau of the Census; (1984) . S tatistic a l A bstract q £ t he 
United States, 1985 (105th Edition) . Washington, D.C. 

U.S. Dept of Agriculture, Food and Nutrition Service; (1980) 

Southeast Asian American Food Habits. FNS-225 Washington, DC 

U.S. Dept of Health, Education and Welfare, Children's Bureau; (1965) 
The nutritional status of American Negroes. Nutr Rev 
23(6): 161-164 

Vermeer, DE; Frate, DA; (1979) Geophagia in rural Mississippi. 
Environmental and cultural contexts and nutritional 
iitplications. Am J Clin Nutr 31:2129-2135 

Vermeersch J.; Hanes, S; Gale, S; (1984) The National Evaluation of 
School Nutrition Programs. Program iirpact on anthropanetric 
measures. Am J Clin Nutr 40 (su£^) : 414-424. 

Villa Dresser, CM; Carroll, MD; and Abraham, S; (1978) Selected 
Findings: Food Consuirption Profiles of White and Black Persons 
1-74 years of age in the US 1971-1974. Advance data from the 
Vital and Health Statistics of the NCHS #21 

Voors, AW; Frank, GC; Srinivasan, SR; W^ber, LS; and Berenson, GS; 
(1981) Hemoglobin levels and dietary iron in pubescent children 
in a biracial community. Pub Health Rep 96 (1) : 45-49 



180 



Waldman, EB; Lege, B; Oseid, B; Carter, JP; (1979) Health and 

nutritional status of Vietnamese refugees. South Med J 72(10): 
1300-1303 

Ward, GW; (1982) Improving public understanding of high blood 
pressure. Urban Health 11(4) : 117-118 

West, KM; (1974) Diabetes in American Indians and otter native 
populations of the New World. Diabetes 23(10) : 841-855 

Willet, WC; MacMahon, B; (1984) Diet and cancer. An overview. N Eng 
J Med 310:633-638 and 697-703. 

Williams, CL; Carter, BJ; Arnold, CB; V^der, EL; (1979) Chronic 
disease risk factors among children: the "Know Your Body' Study. 
J Chron Dis 32: 505-513 

Windham, CT; wyse, BW; Hurst, RL; Hansen, RG; (1981) Consistency of 
nutrient consuirption patterns in the United States. J Am Diet 
Assn 78(6): 587-594 

Windham, CT; wyse, BW; Hansen, RG; and Hurst, RL; (1983a) Nutrient 
density of diets in the USDA Nationwide Food Consunption Survey, 
1977-1978. I. Impact of socio-economic status on dietary 
density. J Am Diet Assn 82(1) : 28-34 

Windham, CT; wyse, BW; Hansen, RG; (1983b) Nutrient doisity of diets 
in the USDA Nationwide Food Consumption Survey, 1977-1978. II. 
Adequacy of nutrient density consunption practices. J Am Diet 
Assn 82(1): 34-43 

P^ant, KW and Meiselman, HL; (1979) USAF Food Habits Study: Part II. 
Food Preferences of Whites and Blacks and Males and Fanales. US 
Army Food Sciences Lab. Tech Rep #TR:-79/042 Natick, Mass. 

Yang, GIP; Fox, HM; (1979) Food habit changes of Chinese perscais 
living in Lincoln, N^raska. J Am Diet Assn 75(4) :420-424 

Yano, K; Rhoads, GG; Kagan, A; and Tillotson, J; (1978) Dietary 
intake and the risk of corwiary heart disease in Japanese men 
living in Hawaii. Am J Clin Nutr 31(7) : 1270-1279 

Yano, K; Wasnick, RD; Vogel, JM; and Heilbrun, liC; (1984) Bone 
mineral measurements among middle-aged and elderly Japanese 
residents in Hawaii. Am J Epi 119(5) : 751-765 

Yanochik-Owen, A; White, M; (1977) Nutrition surveillance in Arizona: 
selected anthropometric and laboratory ctoservations among 
Mexican-American children. Am J Pub Health 67(2) : 151-154 

Yu, MC; Ho, JHC; Ross, RK; Henderson, BE; (1981) Nasoptoryngeal 
carcinoma in Chinese - salted fish or inhaled smoke? Prev Med 
10:15-24 

181 



SOPPLEMENTARy REFERENCES 



Armstrong, H. (1975) Nutritional status of black preschool children 
in Mississippi. J Am Diet Assn 66(5) : 488-493 

Brittin, HC; Zinn, DW; (1977) Meat-buying practices of Caucasians, 
Mexican-Americans, and Negroes. J Am Diet Assn 71:623-628 

Caliendo, MA; (1981) Nutrition and Preventive Health Care. Chapter 5. 
Mac Millan Pub Co, NY Ep.133-169 

Caster, VTO; (1980) The core diet of lower-economic class women in 
Georgia. Food and Nutrition 9:241-246 

Cerqueira, MT; Fry, MM; Connor, WE; (1979) The food and nutrient 
intakes of the Tarahumara Indians of Mexico. Am J Clin Nutr 
32(4): 905-915 

Chang, B; (1974) Some dietary beliefs in Chinese folk culture. J Am 
Diet Assn 65(4) :436-438 

Chase, HP; Kremar, U; Dodds, JM; Suberlick, HE; Hunter, RM; Burton, 
RS; and Spalding, V; (1971) Nutritional status of preschool 
Mexican-American migrant farm children. J Dis Child 122:316-324 

Costas, R; Garcia-Palmeri, MR; Nazario, E; Sorlie, PD; (1978) 

Relation of lipids, weight, and physical activity to incidence of 
coronary heart disease: The Puerto Rico Study. Am J Cardiol 
42:653-658 

Dales, LG; Friedman, (3); Ury, HK; Grossman, S; and Williams, SR; 
(1979) A case-control study of relationships of diet and other 
traits to colorectal cancer in American blacks. Am J Epi 
109(2): 132-144 

Desai, ID; Lee, M; (1974) Nutritional status of Canadian Indians: I. 
Biochemical studies at Upper Laird and Ross River, the Yukon 
Territory. Can J Pub Health 65(5) : 369-374 

Dilling, LA; Ellestad-Sayed, J; Coodin, FJ; Haworth, JC; (1978) 

Growth and nutrition of preschool Indian children in Manitoba; I: 
Vitamin D deficiency. Can J Pub Health 69(3) : 248-252 

Dratanan, RS; Haniner, D; Jarvie, GJ; (1977) Eating styles of obese and 
non-obese black and White children in a naturalistic setting. 
Addictive Behavior 2(2-3): 83-86 



182 



Farris, RP; Cresanta, JL; Frank, QC; Webber, LS; and Berenson, GS; 
(1984) Dietary studies of children from a biracial population: 
Intakes of fat and fatty acids in 10 and 13 year olds (Bogalusa 
study) . Am J Clin Nutr 39:114-128 

Ferris, AG; Vilhjalmsdottir, LB; Deal, VA; and Pellett, PL; (1978a) 
Diets in the first six months of infants in western 
Massachusetts. I. Energy yielding nutrients. J Am Dietet Assn 
72:155-159. 

Gillum, RF; Grant, CT; (1982) Coronary heart disease in black 

populations. II. Risk Factors. Am Heart J 104(4 Pt 1): 852-864 

Guthrie, H.A. (1979) Introductory Nutrition . 4th edition. The C.V. 
Hosby Co., St. Louis, Mo. 

Haider, SQ; Wheeler, SM; (1979) Nutritive intake of black and 
Hispanic mothers in a Brooklyn ghetto. J Am Diet Assn 
75(6):670-673 

HoK^ner, K; McLaughlan, JM; Shah, BG; Thoirpson, JN; Beare-Rogers, J; 
Ellestad-Sayed, F; and Schaefer, 0; (1978) Nutrient levels of 
some foods of Eskimos from Artie Bay, NWT, Canada. J Am Diet 
Assn 73:257-260 

Jerome, NW; (1982) Dietary patterning and change: A continuous 
process. Contenporary Nutrition 7(6) 

Johnston, JL; Williams, CN; Weldon, KIM; (1977) Nutrient intake and 
meal patterns of Micmac Indian and Caucasian women in 
Shubenacadie, N.S. Canadian Med Assn J 116(12) : 1356-1359 

Koh, ET; Caples, V; (1979a) Frequency of selection of food groups by 
low- income families in southwestern Mississippi. J Am Diet Assn 
74:660-664 

Koh, OT; Caples, V; (1979b) Nutrient intake of low-income black 

families in southwestern Mississippi. J Am Diet Assn 75:665-673 

Lowenberg, ME; Lucas, BL; (1976) Feeding families and children: 
1776-1976. A bicentennial stuc^. J Am Diet Assn 68(3) :207-215 

McMurry, MP; Connor, WE; Cerqueira, MT; (1982) Dietary cholesterol 
and the plasma lipids and lipoproteins in the Tarahumara Indians: 
a people habituated to a low cholesterol diet after weaning. Am 
J Clin Nutr 35(4):741-744 

Mettlin, C (1980) Nutritional habits of blacks and whites. 
Preventive Medicine 9:601-606 

Rush, D; Davis, H; Susser, M; (1972) Antecedents of low birth weight 
in Harlem, New York City. Int J Epi 1: 375-377 



183 



Stavig, GR; Igra, A; Leonard, AR; (1984) Hypertension aitong Asians 
and Pacific Islanders in California. Am J Epi 119(5) : 677-691 

Steranermann, G; Haenszel, W; Locke, F; (1977) Epidemiologic pathology 
of gastric ulcer and gastric carcinoma among Japanese in Hav^i. 
J Natl Cancer Inst 58(1): 13-19 

Wagner, PA; Krista, ML; Bailey, LB; Chistakis, GJ, Jemigan, JA; 

Aran jo, PE; Appeldorf , H; Davis, OG; and Dinning, JS; (1980) Zinc 
status of elderly black Americans from urban-low-income 
households. Am J Clin Nutr 33:1771-1777 

Wenkam, NS; Wolff, RJ; (1970) A half century of changing food habits 
among Japanese in Hawaii. J Am Diet Assn 57(1) : 29-32 

Yohai, F; (1977) Dietary patterns of spanish-speaking people living 
in the Boston area. J Am Diet Assn 71(3) :273-275 



184 



i^)pendix 1: 

Background Information on Vitamins and Minerals 



VITAMINS 
Vitamin A 



Vitamin Bl 
(Thiamine) 



Vitamin B2 
(Riboflavin) 



Vitamin B6 
(Pyridoxine) 



Vitamin B12 

(Cyanoco- 

balamin) 



Vitamin C 
(Ascorbic 
acid) 



FUNCTIONS 

Aids in normal growth, 
helps to prevent 
infections, pranotes 
healthy skin, and aids 
in night and color 
vision 



Assists in carbohy- 
drate metabolism. 
Maintains reserves of 
energy. 



Helps to release energy 
from food. Helps main- 
tain healthy skin. 
Assists in the formation 
of red blood cells. 

Assists in the formation 
of the neurotransmitters 
inportant in the brain 
function and in protein 
metabolism. 



Assists in the develop- 
ment of red blood cells. 
Helps maintain nerve 
tissue. 



Forms collagen, a 
substance that holds 
body cells together. 
Hastens the healing of 
wounds. Enhances iron 
absorption. 



IMPORTANT FOOD SOURCES 

Liver, butter, carrots, 
cantaloupes, dark green 
leafy vegetables, 
apricots, broccoli, 
fortified margarine, 
sweet potatoes, cheese, 
cream, & milk fortified 
with vitamin A. 

Liver, meat, poultry, 
whole-grain flours and 
cereals, wheat germ, 
seeds like sunflower and 
sesame, legumes, nuts, 
leafy vegetables. 

Liver, kidney, lean meat 
cheese, milk, yogurt, 
eggs, leafy vegetables, 
beans & peas, fortified 
grain products. 

Muscle meats, liver, 
whole-grain bread, flours 
and cereals, soybeans, 
bananas, peanuts, 
potatoes, beans, brown 
rice. 

Vitamin B12 can only be 
found in animal products: 
meat, poultry, fish, 
shell fish, eggs, milk & 
milk products. 

Citrus fruits, tomatoes, 
cantaloupe & other 
melons, berries, green 
leafy vegetables, peppers 
broccoli, cauliflower, 
fresh potatoes. 



continued 



185 



i^peidix 1: (continued) 

Background Information on Vitamin and Minerals 



VITAMINS 
Vitamin D 



Vitamin E 



FUNCTKMJS 

Promotes healthy teeth 
& bones. 



Prevents cell roenbrane 
damage. 



IMPORTANT POCD SOURCES 

Fortified milk products, 
fish oils, egg yolks, 
butter, liver, fatty 
fish (like sardines, 
salmon, and tuna) 

Vegetable oil, margarine, 
shortening, green leafy 
vegetables, asparagus, 
rice germ, wheat germ, 
butter, liver. 



Vitamin K 



Niacin 



Promotes blood clotting. 



Essential in releasing 
energy from carbohydrates, 
fats & protein. 



Green leafy vegetables, 
liver: Note: Main source 
is synthesis by normal 
bacteria, in the 
intestine. 

Organ meats, lean meats, 
poultry, fish, fortified 
grain products. Nuts, 
seeds, beans & peas. 



MINERALS 
Calcium 



Chloride 



FUNCTIONS 

Present in the body in 
greater amounts than 
any other mineral; 
needed for hard bones 
& teeth, for normal 
behavior of nerves, 
muscle tone & 
irritability, & blood 
clotting. 

Part of hydrochloric 
acid, which is found 
in high concentration 
in gastric juice and 
is important in 
digestion of food in the 
stomach. 



IMPORTANT FOOD SOURCES 

Milk & milk products, 
dark green leafy veget- 
ables (except spinach & 
chard) , citrus fruits, 
dried peas & beans, small 
fish eaten with bones, 
tofu (soybean curd) 



sodium chloride (table 
salt) 



continued . . . 



186 



i^pendix 1 (continued) 

Background Information on Vitamins and Minerals 



MINERALS 
Magnesium 



Phosphorous 



Potassium 



Sodium 



Sulfur 



Iron 



FUNCTICWS 

Found in all body tissues, 
principally in the bones; 
is an essential part of 
of many enzyme systems 
responsible for energy 
conversions in the body 

Is present with calcium, 
in almost equal amounts 
in bones and teeth and 
is an important part of 
every tissue in the body. 

Major constituent of 
fluid inside individ- 
ual body cells. With 
sodium, helps to 
regulate body fluid 
balance & volume 

Found mainly in blood 
plasma and in the fluids 
outside body cells. Helps 
to maintain normal 
water balance inside 
and outside of cells. 

Present in all bo^ 
tissues, related to pro- 
tein nutrition (is a 
coirponent of several 
amino acids) ; also part of 
two vitamins: thiamine 
and biotin. 

An essential part of 
hemoglobin, to carry 
oxygen to body via the 
red blood cells. Lack 
of iron can cause anemia. 



IMPORTANT FOOD SOURCES 

Nuts and beans, whole 
grains, green leafy 
vegetables. 



Meats, poultry, fish, 
eggs, whole-grain foods, 
beans & peas. 



bananas, oranges, dates, 
cantaloupes, tanatoes, & 
baked potatoes, vege- 
tables, meats, poultry , 
fish, milk 



Salt, salted foods, MSG, 
soy sauce, baking powder, 
cheese, processed foods 
such as breads, cereals, 
ham, bacon, crackers. 



Eggs, meat, milk & 
cheese, nuts, legumes. 



Liver, red meats, dried 
beans and peas, enriched 
or whole grain breads and 
cereals, prunes, raisins. 



continued. . . 



187 



Appendix 1: (continued) 

Background Information on Vitamins and Minerals 



MINERALS 
Iodine 



Zinc 



Fluoride 



Copper 



Manganese 



Chromium 



FUNCTIONS 

Forms part of hormones 
of thyroid gland, which 
helps regulate body 
metabolism. Lack can 
cause goiter. 

Needed for tissue repair 
and normal growth of 
skeleton. Part of 
several horomones, in- 
cluding insulin. In- 
volved in cell metabolism. 

Helps prevent dental 
caries. Helps stabilize 
bones, teeth. Deficiency 
can cause tooth decay, 
osteoporosis. 

Vital to enzyme system 
and in manufacturing red 
blood cells. Need for 
utilization of iron. 
Anania possible, but 
deficiences are rare. 

Vital to various enzyme 
systems involved in pro- 
tein and energy metabo- 
lism. Essential for 
normal bone structure 
and functioning of 
central nervous system. 

Essential for normal 
glucose metabolism. 
Helps regulate insulin 
levels. 



IMPORTANT FOOD SOURCES 

Iodized salt, seafood, 
plants grown near sea. 



Red meat, milk, liver, 
seafood, eggs, whole 
grain or fortified cereal 



Fluoridated drinking 
water best source. 



Oysters, nuts, liver, 
kidney, whole grain 
breads and cereals, 
mushrooms. 



Nuts, whole grain breads 
and cereals, tea, vegeta- 
bles, fruits. 



Brewer's yeast, cheese, 
whole grains, meats. 



continued . . . 



188 



i^pendix 1: (continued) 

Background Information on Vitamins and Minerals 



MINERALS 
Selenium 



FUNCTIONS 

Essential role in enzyme 
systems of animals and 
proper functioning of 
blood. 



IMPORTflNT POOD SOURCES 

Varied diet provides 
adequate amounts - fish, 
meat, breads, cereals. 



Molybedenum 



Essential to function of 
enzymes involved in 
production of uric acid 
and in oxidation of sul- 
fites and aldehydes. 



Meats, grains, legxmies. 



SOURCES: National Health Information Clearinghouse "Health Finder: 

Vitamins." Office of Disease Prevention and Health Promotion. 
Public Health Service; U.S. Dept. of Health and Human 
Services. February 1985. 

A Primer on Dietary Minerals. FDA Consumer. Septertoer 1974. 
HHS Publication No (FDA) 77-2070 . 

Lecos, C. Tracking Trace Minerals. FDA Consumer. 
July/August 1983. HHS Publication No (FDA) 83-2176. 

Food and Nutrition Board. Reconmended Dietary Allowances. 
Ninth Revised Edition. National Academy of Sciences. 
Washington, D.C. 1980. 

Mayer J. A Diet for Living. New York. David McKay Caipany, 
Inc. 1975. i^:pendix 5. 



189 



T3 
C 
0) 
P. 
Cu 

< 



o 




z 




3 




O 




(J 




X 




o 




P£ 




<■ 




u 


T 


IZl 


>! 


Si _ 


S) 


, o 






J- 


< 2 

z _ 


-S 


2^ 


-S( 


■a 


h > 




<^ 


«. 


i^ 


cA c« 


•S! 


U u 


? 


u CI 




2Z 




a < 
^1 


u. J 


1 


C -J 


•? > 


^ 


'r^^ 


g 


51= 




< Q 


e 


i > 


■5 


< J 




Z — 


ho 


'^ < 


■^ 


=: Q 




zS 


g 


ii 


.§ 


ni: 


^ 


pa O 




zl:^ 


^ 


O 06 


"! 


h 


^n 






e< 


5 


f- 


Q 


o 




z 




Q 




Z 




< 




a 




Q 




o 





■a 60 



— 


^ 


£_ 


— 


a. 


. 


E 


'Si 








CJ 


u 




c 












F 






5 


e^ 






SI3 


> 


cc 


i 








£ 


CQ 


-1 




c 


bc 



oooooooooooooooino 

+ 



OiniCOOXXOOOOCXccXO-cr-C 



? O »J^. ^ 
— CM ^^^ ^J 






C'^ CO 50 CC 



SC CO CO — — 

+ + 










z 




(J 


&c 




z 


fc. 


R 


c 


"So 


oi 




E 



J? 



h E 

, U 
2 c 

> i 



, w a 

2 c ac 

>E e 

> E 5 

, < s 

> s 6 



Is 



be <u 

< ^ 



oooooooo 

OttoOOOoO 
co_irixxxc*iCMX 



o o o c; o — c: 
c o o o o o o 

X CM CM X X X TT 



lOinomOOooooooooooo 
O — csicMcocOcOcococococoeococo— - — 

+ + 

OiiOOOOOOoOOOOOOOOO 

co-^oooOooooooooooo 
— c^co'^'T'r'^'VTj-Tr'r'^'^-v"- 



co to 



O^cOtOXOCMCMCMXOOOOtD^ 
OOO — — — CMCMCMCM — CMCmCMCMOO 

+ + 



OXiJi— ^DXXO'iXtCiinT-^COcOCMtO 

+ + 

■^^pxo'^^t^r^to-TfiOcocoff^CMco^ 
ooo— '—■•— — «•- — — — — —* — oo 

coirir^^GMTr-^iOTrc^i — — — oo-^^ 
odoo-^ — -^ — — — -; — -; — -^do 

+ + 



COcOTTT'T'iritDtOtOtOiricO'^DtOfOCM'T" 



eo'^intor>-xooooxxxxxc^ico 



^^ 



ooooooooo 
cmoooooooo 



o o o o o o o 

o o o o o o o 

X X X X X CM TT 

+ + 



CM O 
CM CM 
X X 



eoOTpuDtO(OtOtO'OtO'T""*TroO 

o^icocofifivnio^'^'^'V'^TrcocM 



TTXiTlTrCMCMCTiOOOCMTp-TfTf'Tf 

CMC^jco-^iOtotof^r^t^^^totcco 

S— pCMCMr^^Ot-xxr^cocococo 
r>.ai — coinr^r^r^r^iOco=0(0(0 



--cjCM'^toai^unininocMCMCMCM 



tocic^oxintoo 
— CM CM ■^ to r^ 



'^ iTj ifi lc in 

T lO in uO lO 



>n 


o 




— en 
1 1 

lO — 


o 


o 



— ^— c^in — — — C4if) 



So iJ 



5 = = 



2 CO 



= c -■ 



4j ^ n 

> t/l u 

= f 1 

C -J g 

■a ra c 

« 6C 



J3 o il 

""S3. 
5 to "^ 

i; u J: 
« £ 5 



I- - .i 



S .2 



- -z J3 "^ — 



i: Ji 



■3= ~ = i 

>; oa t' i- 

I .SCO 

=- C = Tl 



^ £ I 

« = - 

2 E -a 

£ = 



c c 

C ;_ 



c 2 = 



c ■" 

"^.£ 

bo bO 



wi — :r > 



s -5 



u c 
■o 
c 



c c 
to 

E ii := 

i £- = 
■^ = i= 

DCS 



C 
U .Z 

E r 



= c S ^ 



-a 
■c 

E -0 
.£ c 



E 
£ 

c 



S 60 



■ - B -Z ^ 

I 2 - § i 

I j; — fO w 

; ■£ h - .£ 



■D s 

(U ■ = 

^E 
f- < 



o c 


M 




.h £ 


£ 










2 E 


c 




c ? 


E 




n c 






1^ 


1 


c 


E ■= 


i 


a. 



1.5 
I t 

C EA 

en ^ 
C « 

•2^ 



3 "O 



£ 



i: "o X 






to. 

60 

a. 






X — 



= :2 



■5 i; 

o -o 

V u 

■O ,9 



i3 2 

X C 



° ii E 






« a 
.5 S 

■a -J 

-Q 

60 "O 

£ ^ 



fc 60 O 3 

£? i.E r 



2 3 



3 O 



a. 

c 
•a 

lU -' 

-o -o 

Id 



u -a 

C 4J 



V 



^^ > g 

ii T3 H. 



S II 

2 -O 

^ <ii 

2- a 



o z 

c -a 

u ■" 

« (0 

o t 

60 S 

E ^ 






2 -3 -a t- 



_ -r -r s to 



= * 

», 

Etn 



60 « 
tj 60 E 

u 3 :: 



' jz yy 4; 



= ? 



< a. 

c o 
E ~ 



. c 2 — 

I 2 = t) 

^ '> — ji 

> :t3 - 

■- — > :o 



.£ o 



il 



iS a. rt 



= ■«-=: 



i; ^ ■- "s 



jn 5 c r^ .3- -C 






Z j; 

- h 



190 




Health Care Service 
Delivery in Asian American 
Communities 



^ 



Reiko Homma True, Ph.D. 

Deputy Director 

City and Council of San Francisco 
Community Mental Health Services 
San Francisco, California 



HEALTH CARE SERVICE DELIVERY IN ASIAN AtTERICAN COMMUNITIES 

INTRODUCTION: 

Tne American "nealtn care system nas ccme urder increasing scrutiny during 
tne past decade. Spiralling costs and insensitive, inapropriate , or 
questionable quality of care nave been issues of particular concern. Miorig 
patient groups, tne poor and racial minorities are identified as tne groups 
consistently receiving poor services (DHEW, 1971; Weaver, 1976). In spite of a 
series of legislative initiative witn tne intention of imprcving tne nealtn 
care systan and making it more responsive to tne underserved population, tne 
results so far have not been encouraging. 

When considering services for Asian Americans as a subgroup of tne minority 
canmunity, investigators are often ranirded of tne degree of neglect 
tnatpersists in tne medical canmunity. Tne services available to tnan are for 
tne most part inadequate ard inappropriate. Recently, Asian Jtoerican leaders 
ard concerned nealtn professionals have been working toward cnanging this 
unfortunate situation by involving legislators, goverrment agencies, and 
furding sources. Tnis paper is a r&/iew of the health care issues ard problons 
in tne Asian Anerican canmunity ard successful models and strategies for 
service delivery. 

BARRIERS TO HEALTH CARE UTILIZATIOSf: 

In spite of the visible evidence of increasing health care problons in tne 
Asian Merican canmunity, cursory investigations into health care utilization 
by Asian Anerican have revealed consistently low rate of utilization (Vtong, 
1975; Oriental Service Center, 1970; Li, 1972). Although low utilization is 
often equated witn, or considered related to a low morbidity rate or absence of 
need, studies have shown that there are other factors involved in creating such 
a pattern (Eaton ard V7eil, 1955; Selltiz et al., 1959). Examples of these 
contributirg factors are physical and psychological barriers to service 
utilization, lack of information or availability of alternative resources, ard 
lack of financial resources. 

In the case of Asian Americans, the findings fran canmunity surveys 
identified a variety of factors that limit the accessibility of existirg health 
care services. Following are sane of the more canmonly identified problens: 

1. Language Barrier: 

Altnough Asian immigration to tne United States began nearly 150 years 
ago, ever 60% of the U.S. Asian Anerican population are recent inmigrants . 
While a fair nunber of than are well educated ard are able to deal with the 
danards of tne nav culture ard ne^ larguage, a majority have major problans 
ccnmunicatirg in English. Because of the liberalized immigration law 
enacted in 1965 ard the influx of refugees, tnis trend is expected to 
continue for sanetime. Tnis will meari continued problans for many Asian 
Americans in dealirg with English speaking services. Sirice fev of the 
established health care institutions have bilirgual personnel on hand, 
these services have ranained inaccessible to most of the non-English 
speaking Asian Anericans. 



193 



2. lack of Funds: 

Altnougn there is widespread perception of Asian Jtoericans as an 
affluent and successful minority, the reality is that a substantial nuriber 
are living under marginal circumstances (U.S. Civil Rights Cfcrtmission, 
1980) . Ammong these, inmigrants and the elderly are consistently 
identified as the groups most in need. Although Medicaid and other 
gcvernnent funding have provided sane relief, many do not use these 
resources because they are deeply afraid that the receipt of public 
assistance, be it medical or otherwise, will jeopardize their immigrant 
status. Others are still deeply rooted in the cultural tradition of the 
old world not to seek outside help and to take care of one's needs through 
mutual family assistance. 

3. Location of Services: 

VJhile the majority of poor and elderly Asian Americans reside within 
the boundary of Asian American ghetto ccramunities such as the Chinatowns in 
Nsv York or San Francisco, most health care resources are located in areas 
trat are often difficult for than to reach. The more affluent Asians have 
access to private transporttion, and are able to reach services with little 
difficulty. Others, who must rely on public transportation, have 
difficulty getting to services located outside their area particularly when 
they are rot well. Whenever new services have been located within these 
connunities, the response of the residents has been positive, thus 
affirming the fact that geographic accessibility is a crucial factor for 
service utilization. 

4. Psychological Barriers: 

Mthough more subtle than the concrete problans of funds, or larjguage, 
there is a significant issue of psychological barriers between the Asian 
American consumer and tne health care professionals and institutions. Many 
immigrant patients are rot used to dealing with non-Asian professionals or 
institutions with few Asian personnel present. Although their fears are 
often vague and not necessarily based on reality, many of than have 
encountered mistreatment which could be understood as racian or have heard 
stories of racist treatment. Such fears are shared by other minority 
groups and tend to be pervasive unless an aggressive effort is undertaken 
by the institutions to reach out to tne patient groups. For this reason. 
Whether justified or not, these fears continue among the less acculturated 
Asian Anericans. 

Furthermore, the Eastern medical tradition, familiar to most recent 
Asian immigrants, is very different fron the Western medical tradition 
(Hessler et al, 1975; Chan and Chang, 1975). Most Anerican health care 
professionals are steeped only in tine Western tradition and are not 
sensitive to the traditions annd expectations of their Asian patients. For 
this reason, they are not able to help than develop a trust of Western 
medicine. In contrast, patients feel much more confortable and trusting of 
Asian physicians who urderstand tne patients' need to rely on sane 
established folk medicine and who can help than sort out its limitations 
and the need for Western medical treatment. 



194 



5. Special Healtn Care Needs of Asian Mericans: 

Tne criticism often directed at tne establisned medical institutions, 
even by tnose Asians wno do not have language or financial problans, is 
that tne institutions are insensitive or irdifferent to tne special health 
care needs of Asian Anericans (Weaver, 1976). Following are sane examples 
warranting special attention, which have not yet been addressed at most 
large institutions because they do not consider Asian American groups as 
significant erough svispopulations . 

a. Chinese Anericans: 

According to the health surveys corducted in various parts of the 
Chinatowns in the United States, the following were identified as 
critical health issues: significantly high incidence of tubercialosis, 
hypertension, infant mortality and poor neonatal care, ard problans 
related to the elderly (Huang and Grachow, 1974; Hessler et al., 1975; 
Vtong, 1975; Chan and Chang, 1975). Since these surveys focused on 
residents of Qiinatcwn, where there is a concentration of immigrants 
with marginal resources, they fail to provide an o/erall perspective 
on the needs of the native-born Chinese Anericans ard others who live 
outside of the Chinatown area, ffcwever, the findings are instructive, 
since they identiify the needs of the most vulnerable population in 
the cormunity. Other studies also have identified the terdency among 
Chinese toward psycnoscmatic (Tseng, 1975) ard changing patterns of 
cancer incidence (King and Haenszel, 1973; Fraimeni and Nfeson, 1974). 

b. Japanese Anericans: 

VJHile sane consider tne Japanese Anericans to have a more 
favorable mortality rate than Caucasians (Gordon, 1957, 1967; PCitaro, 
1971 ) , there are no substantiating data for sucn assumptions . Weaver 
(1976) cites a nuriber of existing pathologies tnat plague Japanese 
Anericans more frequently than otner groups, such as a greater 
occurence of cancer of the esophagus, stonach, and liver, and 
psychoscmatic illnesses. Although changes in diet ard lifestyle may 
account for the increasing incidence of various pathologies such as 
cancer of colon, rectun, breast, cardiovascular ard renal diseases, 
little attention has been given to identifying the causes and 
developing preventive strategies. 

c. Filipino Anericans: 

The Filipino catmunity, which is the fastest growing Asian 
canmunity with tne largest nuriber of new immigrants, is experiencing 
a great deal of stress because of the lack of social ard econonic 
resources within itself. Sane of the unique health problens 
experienced ty the Filipino in the United States sean to be related to 
a canbination of genetic predisposition, psychosocial stresses, and 
changing lifestyles. Major health problens include high rates of 
hyperuricemia, particularly among adult males; cardiovascular ard 
renal diseases amorg wanen; diabetes mellitus; ard irdustrial 
diseases. Ailments occurring amorg Filipino farm workers and factory 
workers include arthritis, pesticide poinsoning, and bronchial 
conditions (Weaver, 1976; Cachola, 1984). 

195 



SUCCESSFUL M0DEI5 OF SERVICE DELIVERY: 

Frustrated by tne failure of the establisned healtn care institutions to 
provide appropriate services for tneir own catmunities , a nutiber of concerned 
Asian Merican leaders have acted on tneir own initiative to advocate for ard 
create services they knew to be more acceptable to their people. Beginning in 
the '70s tnis increase of interest acA activity parallel eS tne actions of 
various federal, state and local agencies to facilitate service delivery to the 
poor and minorities. 

When considering service development, the organizers nad to take account of 
tne v/ide variability in the characteristics of each ccmmunity, e.g., size, 
number of etnnic groups present, cluster pattern, professional resources, 
fiscal resources, social-political climate. For tnis reason, tne models that 
eventually emerged successfully ard achieved a level of stability are diverse. 
ffcw&/er, cross cutting all models, several elanents may be identified as 
essential for tne viability of services within the Asian Merican ccmmunity: 

1. Tne services should be located witnin the cotmianity or be as geographically 
accessible as possible. 

2. Tne services should be staffed by bilingual personnel appropriate for the 
population to be served. 

3. Tnere should be an involvenent of ccntnunity menbers in the process of 
program planning, development, and operation. 

4. Tnere should be a provision for sv±)sidizing those without resources at low 
or no cost. 

5. Services offered snould include outreach, supportive services, and healtn 
education, providirjg opportunities to reach the treatment resi stent or 
difficult to treat patients. 

Based on the analysis of the models currently existing in the Asian 
Anerican ccmmunity, tne following models may be identified: 

1. Qanmunity Based, Primary Care Service: 

Because of the relative ease of organizing an outpatient service 
providing primary health care ard tne reasonableness of the start-up cost, 
it is by far tne model most frequently adopted by coimunity groups. Often 
initially startirq as a shoe-string, storefront operation, staffed by a few 
dedicated healtn care professionals, a nunber of them have now become well 
established in their respective canmunities . Deperding on the 
circimstances involved at the time of the prog ran developnent, the 
sponsoring organization coiiLd be private non-profit or local goverrment. 

Cninatcwn Healtn Clinic , located in the heart of Chinatown in New 
York City, is an exanple of such a clinic. It is supported by an Uitoan 
Healtn Initiative Grant, ISfetional Healtn Service Corps, Medicaid, ard a 
small amount of tnird party ard patient fees. Their total budget in 1983 
was $549,000. Respofding to the great need for bilingual services, the 
Clinic provides bilingual primary care services ard healtn education to the 
Chinatown residents and tne surrourding conmunity. 

196 



Tne denograpnic profile of tne clients seen in the Clinic annually are 
as follows: 

-Over 90% of tne Clinic's patients are below 150% of national poverty line. 

-O/er 90% are of Cninese descent, including etnnic Cninese refugees fran 
Soutneast Asian. 

-Over 56% do not nave any sources of support for healtn care, including 
Medicaid and Medicare. 

^any are workers in tne garment irdustry ' s sweatsnops . 

Because of tne extremely lew economic status of tne client population, 
the Clinic is heavily deperdent on gcverrment funding and will ranain so 
for a long period of time. Tne unit cost for a pnysician visit in 1983 was 
$30.95. 

While Chinatcwn Health Clinic is primarily supported by Federal grant 
funds, the Asian Healtn Services (AHS) , establisned in 1974 in Oakland, 
California, is supported by state arid local funds to a significant degree. 
Because of tne presence of a greater variety of Asian groijps in the area, 
the clinic is utilized by several Asian groups. For tnis reason, the 
clinic is staffed by bilingual staff wno speak Cantonese, Mandarin, 
Vietnamese, Korean and Tagalog. Anong tne 2500 patient families wno are 
users of the clinic services, their danograpnic profile includes: 

-60% wno are Cninese; 10.6% Pilipiros; 8.7% who are Koreans; and an 
increasing number of Southeast Asian refugees. 

-Q/er 83% are non-English speaking immigrants . 

-70% are on Medicaid or nave no insurance. 

-Tne nimber of average visits to the Clinic per user was 4.4. 

Like the New York clinic, the Oakland clinic is also funded by Urban 
Health Initiative. In adition, they receive state and local funding, as 
well as otner third party insurers and client fees. Tneir budget in 1984 
was approved at $680,000, whhich was to serve the special needs of the 
Asian residents of C&kland ard surrourjding connunities , wnich nave a 
rapidly increasing Asian population, estimated to be about 100,000. 

In addition to the primary care services, other services provided at 
the Clinic include lab tests, immunization, x-rays, and optometry 
services. Tne staff of 6.43 inlcudes 2.04 pnysicians, an optometrist, 
healtn educator, social worker, and 1.39 health aides. 

Possibly because of tne Civil Service restrictions involved in 
personnel hiring, requiring stricter professional and educational 
qualifications, there are fa\7er goverrment sponsored progrcms. Ifcwever, 
one example of such a program may be found in San Francisco, at the 
Northeast Ifealth Center, #4 , located in Cninatown. It is funded and 
operated by the San Francisco Healtn Department. It pro/ ides free or Icw- 

197 



cost bilingual preventive healtn services, wnich incluae: a well-baby 
clinic, prenatal care, family planning, nutrition counseling, WIC, dental 
care and supplanental food screenifjg. 

Tne Center's budget of $1,670,839 (FY 1983-84) is supported by 
conribined furriing fran federal, state, and city (San Francisco) goverrment. 
Tne 39 nealtn care professionals and paraprofessionals available at tne 
Center incluie priysicians, nurses, nutritionists, healtn educator, ard 
healtn aides. The majority of the clients of the Center are Chinese, with 
utilization rarjging among various programs frcm 60% to 100%. 

2. Ctxmiunity Based Ocmprenensive Ifealth Center: 

Altix)ugh there are many advantages to having a canprehensive health 
center designed specifically for Asian Americans, it requires the presence 
of a large user population and the availability of sufficient funds. Even 
tnen, because of the canplexity involved in organizing a multi-service 
systan, only a hardful of ccramunities have succeeded in establishing such a 
center. Northeast Medical Services (NEMS) in San Francisco is considered 
tne largest ccramunity health center serving a predcminatly Asian population 
in the country. Funded by federal, state ard local gcverrments, NEMS has 
been in operation since 1971, when the federal gcvermient was making an 
effort to provide services in Medically Underserved Areas (MLJA) . Chinatcwn 
in San Francisco was identified at tnat time as one of the MUAs. 

One of the unique features of the center is its use of both clinic ard 
catimunity physicians. For primary care, a registrant has tne freeaon to 
choose a femily pnysician frcm either the NEMS clinic or a group of 
contracted private physicians frcm the ccramunity. 

The proportion of patients receivirg primary care frcm NEMS staff vs. 
contract pnysicians is 80% vs. 20%. The FTE count of NEMS health care 
staff is 17.5 FTE currently, ard includes 4 primary care pnysicians, part- 
time specialty physicians, dentists, an optanetrist, pharmacist, 
nutritionist, ard other allied healtn workers. 

In addition to the primary care services, services available at the 
center include: 4 specialty services (acupunture, urology, podiatry, ard 
allergy services), basic laboratory tests, radiology services, nutrition 
counseling, nursing, pharmaceutical services, dental care, optanetric 
services, social services, health education ard outreach services. Tnrougn 
the 50 participating contract pnysicians, NEMS also provides a host of 
other specialty services off site. Altnough NEMS does not provide hospital 
care, they maintain a coordinating ard referral relationship with several 
hospitals, including the private Chinese Hospital ard the county-operated 
general hospital . 

The annual budget of NEMS for 1983 was $3.1 million, of which $2.1 
million were fran the combined funding of local, state, ard federal 
ccrmiunity healtn center grants. Additional re/enues of $520,000 were 
generated fran tnird-party sources, including private insurarjce, Medicaid 
ard lyfediCare. The patient fees, based on a sliding- fee schedule, broijght 
in an additional $480,000. The NEMS patient population served during the 
year was approximately 23,000, bringing the annual cost per patient to 
$130. 

198 



Ttie cnaract eristics of the patients seen at NEMS are as follows: 

-87% are norf-English speaking 

-98% are Chinese ard etnnic Cninese refi:gees 

-19% are elderly 

-Over 60% are at or belov poverty level 

Tne service area population of tne center, as defined by the City of 
San Francisco, includes tne Cninatcwn area arxi ms approximately 70,000 
residents. Of tiiose 70,000, about 50% are Chinese, ffcwever, because of 
the bilingual, and bicultural capacity of the center, other Chinese in the 
City are also attracted to these services. The potential user pool, for 
this reason, could be as high as 50,000 to 60,000. According to the 1980 
census figures, the Asian Merican population in San Francisco is 147,426, 
which is 21.4% of the city's population. Although other Asian groups 
(Filipino, Southeast Asian refugees, Japanese, Ptoreans) are also known to 
have an acute need for specialized bilingual health services, because of 
the limited resources available, development of services for other Asian 
groups, with the exception of refugee services, has been fragmented and 
limited . 

3. Health Nfeintenance Organization: Chinese Hospital, San Francisco 

While most Asian health service agencies serve a large nunnber of poor 
clients and require a significant amount of public funds for basic sipport, 
a group of Chinese physicians, health care providers and canmunity leaders 
recently made a bold mo/e to establish a privately financed health 
maintenarjce organization. 

Chinese Hospital , in San Francisco, was initially established in 
1925 to deal with persistent discriminatory practices against Chinese in 
the general health care delivery system. It was fourded by broadly 
based canmunity groups involving a diversity of political, religious, ard 
social organizations. Initially requiring furding sijpport from the 
canmunity, the hospital now has a solid fiscal and professional base 
in the canmunity, with a large roster of specialists (Loo, 1984). 

The hospital, located in ttie heart of Cninatown, nas 110 affiliated 
physicians and ever 200 ancially staff. Tne nunber of patients admitted 
during the past year (1984) was 2,350. O/er 95% of the patients are 
Cninese, of whon nearly 80% are Medicaid ard Medicare eligible, and the 
majority are rKDn-English speaking. The annual operating budget of the 
hospital is approximately $10 million dollars; the daily bed rate is $530. 

Recently, the Chinese canmunity leaers have been concerned about the 
growing trerd amoung anplcyers, includifjg those of the workirjg class 
Chinese, to purchase econanically more advantageous health maintenance 
organizations (HMO) or preferred provider type health plans for their 
workers. This could mean that many of the working monolingual Cninese will 
be deprived of access to local bilingual health care services. In order to 
forestall further erosion of the services in tne Cninese canmunity, the 
health care leaders in the cannuriity have established an exclusive health 
provider organization of their own. The Chinese Canmunity Health Plan, 
established in 1984, is a collaborative effort of Chinese Hospital, the 
CalifiDrnia Physicians Insurarce Corporation, and ever 70 physicians. 

199 



Tne developmental cxDst for the Plan nas been ever $250,000. In order 
to becane ccmpetitive witn tne larger, more establisned HMCs in the area, 
each of the partners in the venture nas agreed to accept lower 
reimbursanent rates. Tne monthly manbership rates are: individual - 
$62.80; two party - $125.10; family - $179.90. The eligibility for the 
plan is limited to tne working anployees of policy holder organizations. 
Although the plan will nDt be able to assist the unemployed poor, it is 
targeted toward l0A?er income working anployees by keeping its praniun at 
the lowest available rate. Tnis is a bold and new attanpt at meeting the 
needs of working Asians whose resources are limited and their needs 
sonewhat different from those of ti\e general population. 

4. Multi-Service Model: 

While other minority groips have touted the value of locating a 
variety of hunan ard health services under one agency or one lcx::ation 
(Padilla et al., 1975), only a fev Asian Anerican conmunities have 
succeeded in creating such a system. Boston's SoiJth Oove Oatimunity Ifealth 
Center is a successful exception (Lee, 1979). Also located in Qiinatcwn, 
the Center was established in 1972. It provides rot only primary care 
services, dental, eye, nutrition, and health education, but it also has a 
well organized mental health component. This strategy was based on tine 
belief that tne scmatizing terdency of Asian Americans cannot be properly 
treated without coordination between health and mental healtn. In order to 
provide one stop assistance to multi-protolan Chinatown residents, the 
Center works closely with other service agencies under one roof, i.e., 
programs offering housing assistance, vocational trainirg, legal services, 
social services, day care ard youth prograns. Funding for the health 
center is largely public, coming from federal, state, ard city gcverrment. 

Tne Center's budget for 1984 was approximately 2 million dollars. Ttie 
user popxiLation is well over 10,000 and the encounter visits over 40,000. 
O/er 90% of the patients are Chinese ard ever 70% are ron-English 
speaking. In addition to the 3.75 FTE physicians, there are approximately 
14 FTE allied health professionals and ancillary service staff within the 
health unit. 

5. Services for Special Groips: 

Although special needs were identified for the elderly, wanen, or 
youth groups, it is more difficult to d&/elop categorical healtn programs 
designed especially for these groips because of the small numbers in target 
peculations. HDw&/er, possibly because of the cultural value placed on the 
care of the elderly, there seams to be st^jport for service developnent for 
tnis groups in various Asian conmunities. Among tne existing services. 
On Lok Senior Health Services is the most ambitious and ccmprenensive 
systam in the country (Lew, 1984). 

Establisned in 1972 in San Francisco as a non-profit, free-standing, 
it is a ccmmunity-based long-term organization serving the frail elderly. 
Vfi-tn a client population of approximately 300 elderly, about 75% of wham 
are Chinese, the agency is supported by the budget of approximately $1.7 
million. Although they were previously supported by the A3ministration on 
^ing (AC?\.) and the Health Care Financing Aininistration (HCFA) as a 

200 



research and danonstration project, tne agency took a new finarcial 
position in 1984 as a demonstration project. It assuned tne total 
financing risk of providing for all iiealtn services for tne registered 
patients. Tnis danonstration project will assess the feasibility and 
desirability of provider assunption of risk as a means of imprCT/ing the 
quality and controllirjg tne costs of lorg-term care. 

On Lok has ctotained service waivers fran Medicare and Medicaid with 
research ard develcptient funding frcm a consortiun of foundations: Robert 
Itood Johnson, Hartford, Kaiser Family and Retiranent Research. The new 
venture attanpted by On Lok is quite unique in that it is the only 
organization in tne country to assune financial risk, under a fixed 
capitation rate, for an all-inclusive range of services provided to an 
elderly population. On Lok's capitation rate is $1400 per month. For 
Medicare-Medicaid eligible participants, Medicare's monthly share is $650, 
and Medicaid's is $750. For non-Medicaid participants, a copayment systan 
involves participants and family manbers in financing tne cost of care. 

The services offered by the agency include acute hospitalization, 
nursing hone, pharmacy, contracted professional services such as dentistry, 
optanetry and all tne subspecially medical services. On Lok operates 
three adult day health centers ard was the first such program in 1978 to be 
designated as eligible for Medicaid benefits. In addition, they manage On 
Lok HDuse, a HUD developnent of fifty-four apartments housing 66 
ifdividuals, and 6 camiunal living facilities housing 3 to 6 elderly 
residents person each. Also at the site, there is a 2-bed respite unit for 
tne clients of On Lok programs so that those needing tanporary attendant 
care may be acconodated . The unit is atterded by health workers 24 hours a 
day, with easy access to nursing and medical staff. Hone health services, 
transportation, ard nutrition programs coniplete tne range of services 
pncvided. 

Although tne example cited is a program targeted for tne elderly 
population, other specialty services t^^e also been developed in various 
ccmninities . Although not as extensive as the On Ick model, they are 
designed to meet the special needs of particular groups, such a wonen' s 
health, stibstance abuse prograns ard youth progrcBns. Depending on the 
resources available in tne conmunity, it may be more feasible to de/elop 
such specialty services than to de/elop a more ccmprehesive program. 

PROBLEMS IDENTIFIED BY EXISTING SERVICES: 

Many of the Asian health services initiated during the past several years 
have been successful in cwerconing a nuriber of problans identified in earlier 
pages as barriers to Asian service utilization. Hcwever, all of the programs 
cited have indicated their concern that a greater effort must be made to deal 
with a variety of additional obstacles. Following are seme of the more cotmon 
issues: 

1. While the nunber of Asian inmig rants continues to grow, increasing both in 
size and the nunber of ethnic groups represented, the resource allocation 
has not shifted according to the changing nature of the population. The 
programs, therefore, are forced to extend already over- stretched resources 
to meet tne need of the new populations or to ignore their needs. 

201 



2. Tne current systen of funding and reimbursanent , whicn is leaning more 
toward capitated costs, does not allow for tne cost of translation or 
outreach to work with tnese difficult, multi-problem populations. 

3. Current Immigration and Naturalization Service regulations, penalizing 
immigrants fron seeking any kind of public assistance, including medical 
care, are ccmpounding tne dilanma of the struggling irttnigrant groups. 

4. While it is more feasible to develop bilingual services for the more 
doninant, larger Asian groups, it is much difficult to de/elcp resources 
for analler Asian groups, whose individual needs may be just as great. 

5. Although there is a definite trend toward prepaid health care systems in 
the health care industry, specialized ethnic services are lost in the 
process by larger organizations. At the same time, because of their anall 
size, it is difficult for most Asian health care organizations to develop a 
canpetitive prepaid syston. 

6. While many of the existing Asian health care providers are finding the 
advantages of preventive programs in reducing the utilization of more 
expensive services, the reimbursement policies of gcverrment and private 
industries is to eliminate such costs. 

7 . While there is a significant concentration of bilingual Asian health care 
providers in certain metropolitan cities, it is very difficult to recruit 
such personnel in areas with fe^er amenities and a analler number of Asian 
residents . 

8. While an area may have an adequate number of non-Asian physicians, the lack 
of appropriate bilingual physicians could qualify that are an MUA for the 
Asian population. Hcwever, because of the simplistic application of the 
federal regulations, the non-English speaking patient groups are 
underserved . 

RBCCMMEasrmT IONS : 

Based on the experience gained fron various Asian conmunities , we are now 
beginnifjg to understand seme of the essential ingredients for the successful 
health services within our cormunities. Because of the limited resources 
available in tne Asian i^erican carrounity, much support is still needed fron 
federal, state and local gcvernnents. As this paper was developed for the 
purpose of review by the Task Force on Asian Health under the auspices of the 
Department of Health and Human Services, the following recormerdations are 
being made to the l^sk Force for consideration: 

1. Federal support should be sought for the allocation of funds to implonent 
Asian health services which are culturally and linguistically accessible to 
the comiunity. 

2. Federal mardates should be developed to require bilingual personnel in 
health care agencies serving Asian clients with limited Efjglish speaking 
ability. 



202 



3. The Federal Gcverrment, particularly DHHS, snould prcvMe support and 
incentives to tne innovative service delivery models currently in trie Asian 
Anerican cotmunity. 

4. Tne federal gcverrment snould d&^elop mandates to tne states so tnat 
federal furds being administered by tne states will be appropriately 
allocated for program developnent in tne Asian Merican conmunity. 

5. Federal grant support snould be provided to condix:t research into tne 
feasibility of innovative Asian "nealtn services. 

6. Federal support snould be provided to encourage the training of Asian 
health care professionals in both Eastern and Western health care 
practices . 

7. Danonstration Centers snould be established by federal funds to provide 
technical assistance to cotmunities with urderdev eloped Asian health 
resources . 

8. Federal regulation snould be revised to allow flexibility for tne special 
needs of Asian pcpixLation, e.g., MUA description criteria, reimbursatient 
rates, allowable service modalities. 



203 



REFERENCES 



Asian Health Services. Progress Report : Caklarjd, CA: 1984. 

Cnan, C. and Chang, J. K. "Tne Role of Cninese Medicine in Nsa? York City's 
Chinatown," An. J. of Cninese Medicine , 4, 1, 31-45, 1975. 

Cninatown Ifealtn Clinic. Needs-Demand Assesanent . New York: 1983. 

Ccmmission on Civil Rights. Success of Asian Anericans : Fact or Fiction . 
Vfeshirgton, D.C.: 1980. 

Easton, J. and Vfeil, R. J. Culture and Mental Disorders . Glencoe, 111.: 
Tne Free Press, 1955. 

Frauneni, J. F. and IVbson, T. J. "Cancer M^rtiality Anorg Chinese Anericans, 
1950-69," J. of tne National Cancer Institute . 52, 3, 659-665, 1974. 

Gordon, T. "Mortality Experierce Anong the Japanese in the U.S., Ifewaii and 
Japan," Public Ifealtn Reports . 72, 543-553, June 1957. 

Hessler, R. M., Nolan, M. F., Ogbrue, B. and New, P. K. 

"Intraetnnic Diversity: Health Care of tne Chinese Anericans," 
Hunan Organization . 34, 3, 253-262, F^ll 1975. 

Huarg, C. and Gracnow, F. Tne Dilenma of tfealth Services in ChinatOArn , 
New York City . New York City Department of Ifealth. 1974. 

King, H. and Haenszel, W. "Cancer Mortality Anong Foreign and Native Born 
Cninese in the U.S.," J. of Chronic Disease . 26, 623-646, 1973. 

Lee, E. "Mental Health Services for tne Asian Anericans: Problans and 

Alternatives." In Gtsrtnission Civil Rights. Issues of Asian and Pacific 
Americans . 734-757, 1979. 

Li,F. P. "Chinese Cfcntnunity Health T^sk Rjrce Study," An. J. of Public 
Health . 62, 4, 536-539, ;^ril 1972. 

loo, C. M. arxd Yu, C. Y. "Pulse on San Francisco's Cninatown: Health Service 
Utilization and Health Status." Anerasia Journal . In Press. 

LcM, H. On-lok Senior Health Services - Ccmmunity-Based Long-Term Care. 
Unpublished Paper. 1984. 

North East Medical Services. Health Currents. San Francisco: 1984 

Oriental Service Center. Health Surveys . los Angeles, i^ril 1970. 

Padilla, A. M., Ruiz, R. A., and Alvarez, R. Canmunity Mental Healtn Services 
for the Spanish Speaking/Surnamed Population. Anerican Psycnologist , 
30, 892-905, 1975. 



204 



REFERENCES (cx3ntinued) 



Selltiz, C. Jano5a, M., Deutcn, M., and Cock, S. 

Researcn Metnods in Social Relations . New York: Holt, Rine'nart, and 
Winston, 1959. 

Starr, P. Tne Social Transformation of Merican Medicine . New York: 
Basic Books, 1982. 

Tseng, W. S. "Tne Ifeture of Scmatic Cfcmplaints ?morjg Psycniatric Patients: 
Tne Qiinese Case," Oomprenensive Psycniatry . 16, 3, 237-245, 1975. 

U. S. Dept. of H. E. W. Ttwards a Ocmprenensive Ifealtn Policy for the 

1 970 ' s . A White Paper . Wasnington, D.C. : U. S. Goverrment Printing 
Office. 1971. 

Weaver, J. L. National Health Policy and trie Underserved: Etnnic 
Minorities, Wjnen, and the Elderly . St. louis: Mosby, 1976. 

Vtong, M. A Survey Report on Medical Care in Oakland's Asian Ccnmunity. 
Oakland, CA: Asian Health Services, I^y 1975. 



205 



Asian-White Mortality 
Differences: Are There 
Excess Deaths 



Elena S. H. Yu, Ph.D. 

Associate Professor of Sociology in Psychiatry 
and Research Associate 

Ching-Fu Chang, M.S. 
Graduate Research Assistant 
Department of Sociology 

William T. Liu, Ph.D. 

Professor of Sociology and Director 

Stephen H. Kan, Ph.D. 
Research Associate 

Pacific/Asian American Mental Health Research Center 
University of Illinois at Chicago 
Chicago, Illinois 



ACKNOWLEDGMENT 



The authors are grateful to Phyllis Flattery for her helpful 
suggestions and encouragement, and to Aiko Igarashi for her timely 
assistance. Paul Kurzeja's and Medy Masibay's tremendous patience in typing 
the tables in this report is also appreciated. 



208 



ASIAN-WHITE MORTALITY DIFFERENTIALS: 
ARE THERE EXCESS DEATHS? 



Asian Americans are the fastest growing segment of the U.S. 
population today. According to the 1980 Census, the number of persons who 
originate from the Asia/Pacific Triangle has ballooned by 120 percent over 
the past decade — to 3.5 million — while the number of whites has grown by 
6.4 percent, blacks by 17.4 percent and Hispanics by 60.8 percent. 
Factors accounting for most of this increase are: immigration, births, and 
inclusion of new groups in the census definition. The term Asians 
comprises a number of diverse groups; in many ways, they are as different 
from one another as they are from other races. 

More than 20 Asian populations were reported in the 1980 Census. 
Table 1 shows the number and percent distribution of some of these groups 
in 1980 compared with 1970. Chinese, historically the first Asians to 
enter the United States and the first group to be legally barred from 
becoming U.S. citizens (by the Exclusion Act of 1882) , emerged as the 
largest Asian population in 1980. The Japanese, who were the largest 
group in 1970, fell to third in 1980, surpassed by Filipinos, who were the 
second largest group in both 1970 and 1980. 



Table 1 



With the passage of a law in 1981 which allotted a separate 
immigration quota of 20,000 persons per year for Taiwan, in addition to 
the 20,000 assigned to Mainland China by virtue of the 1965 Amendments to 
the Immigration and Naturalization Act, Chinese population growth in the 
coming decades will accelerate more rapidly than that of any other Asian 
group. Yet, except for an overview of the occupational transition made by 
Chinese Americans over the last one hundred years (King, 1981) and a few 
pieces of work based on analysis of 1960 data (Kitagawa and Hauser, 1973; 
King, 1974) , little has been published on the health of Chinese Americans 
or, for that matter, Japanese and Filipino Americans. Indeed, an 
examination of the three most current bibliographies on Pacific/Asian 
Americans (Yu, Murata and Lin, 1982; Doi, Lin and Vohra-Sahu, 1981; 
Vohra-Sahu, 1983) reveals a dearth of research on this subject. 



I. Objectives 

The purpose of this paper is to provide a description of the 
mortality patterns of Chinese, Japanese, and Filipinos in America, by 
examining available data extracted from death certificate records 
submitted by each of the 50 states to the National Center for Health 
Statistics (NCHS) . Created in 1960, the National Center for Health 
Statistics is mandated to collect, analyze, and disseminate statistical 
and epidemiologic data on the health of the nation. However, because the 



209 



size of the Asian American population remained numerically insignificant 
until recently, national data for this special population are difficult to 
analyze and interpret even though they have existed for some time at NCHS 
in two of its vital statistics systems, U.S. birth and death files. 
Furthermore, since analysis of such data depends heavily upon the 
availability of population denominators supplied by the Bureau of the 
Census, the absence of intercensal estimates for Asian Americans in 
general and Chinese, Japanese, and Filipinos in particular, has severely 
limited the use of these records for research purposes. This paper 
focuses exclusively on the mortality patterns of the three largest Asian 
groups in 1980 and compare their differences with the majority white 
population, as well as examines the intra-ethnic differences by age, sex, 
and nativity. In what follows, the term Asians refers only to these three 
groups, unless otherwise specified. 



II. Geographic Distribution and Sociodemographic Profile 

Although the Asian population was more geographically dispersed in 
1980 than in 1970, it remained highly concentrated in the West. In 1980, 
56 percent of the Asian population lived in the West compared with 70 
percent in 1970. However, the degree of concentration in the West varied 
among the groups. For instance, about 8 out of 10 Japanese, 7 out of 10 
Filipinos, but only 1 out of every 2 Chinese were residing in the West in 
198 0. (In contrast, only about 4 out of 10 Koreans and 2 out of every 10 
Asian Indians resided in the West.) 

Regardless of their regional distribution, Asians as a whole lead a 
predominantly urban existence. Ninety-seven percent of the Chinese liye 
in urban areas, followed by Filipinos and Japanese, both 92 percent, 
compared to 71 percent for the majority whites. 

Ferhaps the most impressive sociodemographic characteristic of the 
Asian populations is their level of education. Fully one-third of Asian 
adults have finished college, compared with 17.5 percent of majority 
whites. Further breakdowns by age and sex (Table 2) show that, in the 
male 25 to 35 age group, 57 percent of the Chinese have four years of 
college education or more, compared with 25 percent for whites. The 
corresponding figures for Japanese and Filipinos are 49 percent and 33 
percent respectively. Among females in the same age group, 46 percent of 
both Chinese and Filipinos, as well as 40 percent of Japanese, have such a 
high level of education, compared with 22 percent of whites. The 
educational advantage is sustained in the next higher age groups, 35-44 
and 45-64 years. Among males, almost twice as many Asians aged 35-44 
years (55 percent of Chinese, 49 percent of Japanese, and 48 percent of 
Filipinos) as whites (26 percent) have at least a college education. For 
females, more than three times as many Filipinos (54 percent) and twice as 
many Chinese (34 percent) have such a level of education compared to 
whites (16 percent) . The corresponding figure for Japanese women (25 
percent) is not as strikingly different as those found for the other Asian 
groups. In the 45-64 age range, only Japanese females have lower 
percentage (8.9) of population completing 4 years or more of college, 
compared to white Americans (9.6 percent). 



210 



Table 2 



However, the unparalleled educational attainment of Asians does not 
necessarily assure the advantage that one would expect in the U.S. 
occupational structure. The 1980 census sample data show that, for both 
sexes, 19 percent of the Chinese, 15 percent of Japanese, and 14 percent 
of Filipinos have a professional occupation compared to 13 percent for 
whites. At the other end of the occupational ladder, 19 percent of the 
Chinese, 17 percent of Filipinos, and 13 percent of Japanese are in 
service occupations, compared to only 11 percent of whites (Table 3) . 

Within each Asian group, proportionally more foreign-born persons 
than native-born ones are in service occupations (21 percent for Chinese, 
20 percent for Japanese, 17 percent for Filipinos, compared to 11 percent 
among white Americans) . Although proportionally fewer Asians than whites 
are employed as operators, fabricators, and laborers, more foreign-born 
persons than native-born among the Chinese and Japanese populations are 
engaged in such types of work. Filipinos seem to be an exception insofar 
as nativity is concerned. The proportion of foreign-born persons engaged 
in professional occupations is more than twice that found for the native 
born. This striking contrast is undoubtedly due to the large influx of 
physicians, pharmacists, nurses, and other professionals from the 
Philippines for several decades. 



Table 3 



Income data from the 1980 census reveal that regardless of nativity, 
the median family income for Asians is higher than that found for whites, 
the percentage of families earning $25,000 or more in 1979 being higher 
for Asians than for white Americans (Table 4). However, caution is 
warranted in interpreting such types of data because they mask the true 
poverty situation of Asian Americans. Differences in the family structure 
and kinship concept between Asians and whites enable large numbers of 
Asian adults to tolerate the sharing of a household for reasons of 
exigency, thereby inflating the reported "family" income. Using the 
5-percent sample from the 1980 Census Public Use Microdata, Kan and Liu 
(1984) found that besides the Vietnamese, Chinese and Korean Americans 
actually have poverty prevalence well above the national level. 



Table 4 



Nevertheless, it is clear that changes in the immigration laws and 
sociopolitical developments in the international scene in the last 40 
years have led to the influx of a new generation of Asian immigrants — one 
recruited from a different social stratum in their home country than those 
who came in the 19th century or the first half of the 20th century. Such 
marked shifts in the characteristics of the immigrant population would 



211 



suggest a different mortality (and morbidity) pattern compared with the 
earlier cohorts of immigrants. In the next section, we will examine the 
health data compiled from the birth and death files maintained by NCHS. 
We begin first by reviewing the most recent findings on inter-ethnic 
differentials in infant mortality, followed by general mortality. Next, 
gender differences are compared between groups and within each Asian 
group, using the 1979-1981 mortality files. Finally, because three-fifths 
of the Asian/Pacific Islander population as a whole are foreign-born, we 
will take a look at the significance of nativity, if any, in the mortality 
patterns of Asian Americans. 



III. Infant Deaths 

Infant mortality rates, that is, deaths under 1 year of age per 
1,000 live births, have often been used as an indicator to compare the 
health of different populations. Data on infant deaths by specific 
nonwhite races have been available at the National Center for Health 
Statistics for a number of years. Perhaps the most recent analysis of 
inter-ethnic infant mortality data which include information on Asian 
Americans is the one conducted by Yu (1982) . Using both published and 
unpublished data prior to 1980 on Chinese, Japanese and whites, the author 
compared four different types of rates: fetal (at least 20 weeks of 
gestation) , neonatal (within 28 days of birth) , postneonatal (within 28 to 
365 days of birth) , and infant mortality (Table 5) . She found large 
inter-ethnic differences in fetal mortality (5.9 per 1000 total births for 
Chinese, 8.5 for Japanese, and 9.7 for whites). Likewise, substantial 
differences exist between Asians and whites in the neonatal death rates 
(3.7 per 1000 live births, 5.5, and 10.3, respectively) and in the infant 
mortality rates (5.5 per 1000 live births for Chinese, 7.4 for Japanese, 
and 14.1 for whites). Just why these rates are so different is difficult 
to interpret because of the limited information on the death 
certificates. We may, however, evaluate some possible record-keeping 
errors such as underreporting, misreporting of race, or misclassif ication 
of time of death, and maternal factors such as age and education. 



Table 5 

A. The Possibility of Underreporting 

Since infant mortality rate is defined as: 

the number of deaths in a year of children 

less than 1 year of age x 1000* 

the number of live births in the same year 

two possible sources of reporting error exist. The first lies in the 
denominator which reflects the number of live births; the second may be 
found in the numerator which shows the number of infant deaths. Generally 
speaking, under registration is relatively common among births occurring 
outside of hospitals. However, findings from unpublished NCHS data for 



212 



the entire United States during 1973-77 indicate that only 0.7 percent of 
all reported live births to Chinese as well as to Japanese mothers had 
occurred outside of the hospital setting. The rate for Filipinos, though 
somewhat higher — 1.5 percent — is still low. 

If the underreporting error lies in the numerator, then a systematic 
study is warranted of the death registration procedures — especially in 
those States where these three Asian groups are found in large numbers — in 
order to determine the magnitude of underregistration. Unfortunately, 
such a study has yet to be conducted. 

Relevant to the above issues is the possibility that a vital event 
may not have been registered in both the numerator and the denominator. 
McCarthy et al. (1980) , for instance, demonstrated from their research in 
Georgia that such blatant omissions can indeed happen, although they seem 
to occur disproportionately in the rural areas, among unmarried mothers, 
and for black American infants. In contrast, we know that Asian vital 
events are more likely to occur in urban areas because of the nature of 
Asian settlements in the United States. Compared to either black or white 
Americans, the proportion of unwed mothers thus far is rather low among 
Asians in general and Chinese in particular. Nevertheless, taking the 
McCarthy findings into consideration, we must bear in mind that for 
underregistration of deaths and/or births to be a major source of bias in 
the observed Asian infant mortality rate, we not only need to provide 
evidence that underreporting exists, but we also need to demonstrate that 
the magnitude of underreportage is greater for Asian than for white 
American infants. In the absence of any systematic study in this respect, 
we can neither establish nor dismiss it as a source of error. We can, 
however, momentarily sidestep this issue so that we may proceed to 
evaluate other possible explanatory factors. 



B. Misreporting of Race 

To date, two studies have documented the existence of racial 
misclassif ication in births and deaths for Asian Americans. The first 
uses the 1965-67 linked birth and infant death records in California 
(Norris and Shipley, 1971) to determine if the race of infants born to 
Asian parents are recorded consistently at birth and at death. Norris and 
Shipley reported that the death rates for both the Chinese and the 
Japanese appeared to be substantially lower than the rates for the white 
population, mainly because 39.2 percent of the Japanese infants and 13.7 
percent of the Chinese infants born during that period were mistakenly 
classified as white upon their deaths. Adjusting for this error, they 
found that the Japanese, instead of having a more favorable pregnancy 
outcome than white Americans, actually had higher infant mortality rates. 
However, the Chinese cohort rate, especially for the neonatal period, 
remained to be the lowest of any racial group , despite the adjustment for 
misreportage. 

The second study on racial misclassif ication uses the 1968-77 data 
from Washington State (Frost and Shy, 1980) . Misclassif ication errors in 
Asian births and deaths were found to occur. However, the number of 



213 



Asians living in the State of Washington, especially Chinese Americans, is 
extremely small. Patterson's advice (1980) to take caution in the use and 
interpretation of vital records for numerically small racial groups is, 
therefore, very well taken. 

One finding that emerged from Frost and Shy's study is that 
offspring of interracial marriage appear to have a significantly higher 
rate of discordance between birth and death certificates. However, 
interracial marriage as a factor in the discordance between birth and 
death certificates, though relevant, may not be a truly significant factor 
in accounting for the low Chinese infant mortality rates, although it may 
be significant for Japanese and Filipinos where proportionately more 
interracial marriages have been reported, both in the 1980 and the 1970 
census. For these reasons, the analyses that follow will focus 
exclusively on the Chinese-white differences. 



C. Misclassif ication of Time of Death 

A third possible source of error which might account for the low 
Chinese infant mortality rates is that the time of death may have been 
misreported. For example, neonatal deaths, especially in the first day of 
life, may have been misreported as fetal deaths. If so, we would expect 
to find an unusually high fetal death rate for Chinese compared to white 
Americans. Bearing in mind the uncorrected error of possible misreporting 
discussed above (i.e., some differences in racial classification between 
birth and death records which we cannot establish nor rule out) , we recall 
that in Table 5, the Chinese fetal death rate (5.9 per 1000 total births) 
is actually lower than that found for whites (9.7). Indeed, data 
published elsewhere (Yu, 1982) show that for both fetal and postneonatal 
periods, the rates for Chinese are the lowest of all racial/ethnic groups. 



D. The Importance of Record Linkage 

Frost and Shy's recent publication (1980) and Norris and Shipley's 
earlier work (1971) demonstrate the importance of matching birth and 
infant death certificates in order to approximate the true infant 
mortality rate for different ethnic groups. Since birth and death records 
in the U.S. are not linked nationwide, Yu (1982) examined the data from 
California — where record linkages exist and where 40 percent of live 
births to Chinese mothers occurred during 1973-77. To minimize annual 
chance fluctuations, the linked data we obtained from California are 
cumulated over 5 years (1973-77) . In addition, data for white Spanish 
minorities are differentiated from those for white non-Spanish majority. 
Consequently the observed infant mortality rates for white Americans are 
not artificially elevated by the higher rates found for white Spanish 
minorities. 

The results (Table 6) show that only in the fetal period is the 
death rate for Chinese significantly lower than that found for white 
non-Spanish Americans. Insofar as the neonatal death rate is concerned, 
the observed difference between Chinese and white Americans is not 



214 



statistically significant. In the postneonatal period, the inter-ethnic 
difference disappears. If we take these findings to reflect the true 
rates, then, it seems that some unknown advantage in fetal survival exists 
for Chinese Americans, but this possible advantage is not strong enough to 
be sustained through the first 28 days of life, and it disappears 
thereafter. 



Table 6 



E. Mother's Age and Fetal Deaths 

Insofar as fetal death rates are concerned. Table 7 demonstrates 
that the Chinese advantage is evident in every age group, from 20 to 39 
years. From the magnitude of the observed differences in rates by age, it 
appears that factors other than mere reporting or classification errors, 
are operating. Just what these factors are remains to be explored. 



Table 7 



F. Mother's Age and the Distribution of Total Births 

Another way of gaining insight into the nature of the differences in 
the fetal death rate between Chinese and whites is to look at the age 
distribution of the mothers at the time of the expulsion of the fetus, 
i.e., birth. An interesting finding from the California data is the 
differences in the distribution of total births (fetal deaths plus live 
births) between Chinese and white non-Spanish American women (Table 8) . 
Teenagers accounted for 14 percent of the total births among white 
non-Spanish women, but only 2 percent among the Chinese. Likewise, a much 
smaller percentage of Chinese women (19 percent) gave birth between the 
ages of 20-24 years, compared to white non-Spanish women (35 percent). By 
far the largest proportion of Chinese women — 47 percent — gave birth while 
they were between the ages of 25-29 years, or slightly older, 30-34 years 
(23 percent) . It is noteworthy that the proportion of births to Chinese 
women 35 years of age or over (8 percent) is twice that found for whites 
(4 percent) , Given the knowledge that older women have a very high risk 
of fetal death, the age distribution for Chinese is in fact less favorable 
for fetal survival than that for whites — and yet their fetal death rate is 
much lower than that of white Americans. 

Table 8 



215 



G. Mother's Education 

Given the high level of education among Chinese women in the United 
States, might not the observed low perinatal rates for Chinese be 
accounted for by the mother's high educational level? Unfortunately, 
California's birth certificates do not contain information on mother's 
education. Certificates used in other states with large Chinese 
populations (e.g.. New York and Hawaii) do have information on mother's 
education but tabulations of linked data for Chinese are not routinely 
available. However, since the California data show that the advantage, if 
any, of Chinese appears to exist most distinctly in the fetal period, the 
author decided to examine the 1973-77 U.S. fetal death files using the 
race of mother alone to determine the race of child. The advantage of 
using the fetal death records is that it eliminates the possibility for 
racial discordance between birth and death records, since fetal deaths are 
reported on only one record. 

Table 9 shows the fetal death rate by mother's education. Chinese 
women have a lower rate at every level of education. Thus, the higher 
education of Chinese mothers vis-a-vis that of white mothers does not 
explain the low fetal death rates observed among Chinese. Rather, these 
findings suggest that the explanatory factors probably lie in the maternal 
intra-uterine environment which is, of course, highly influenced by the 
mother's condition and health habits, besides as yet unknown genetic 
factors. 



Table 9 



Current research on spontaneous abortion and perinatal development 
in the biomedical disciplines have succeeded in pinpointing certain key 
factors as critical variables in explaining the fetal death differentials 
observed in non-Chinese populations. In a number of well-designed and 
carefully controlled studies, both maternal cigarette smoking and alcohol 
consumption, for instance, have each been shown to have a dose-response 
relationship to fetal development. Over 200 published works on the 
perinatal effects of maternal cigarette smoking have been reviewed in the 
Surgeon's General Report on Smoking and Health . Therefore, we will only 
recapitulate the important points below. 

First, smoking during pregnancy is a risk factor for spontaneous 
abortions or fetal deaths and perinatal mortality. The highest perinatal 
mortality risk ratios (smokers versus nonsmokers) reported in the 
literature is 2.42. Second, the proportion of pre-term live births 
increases directly with the quantity or frequency of maternal smoking. An 
estimated 11 to 14 percent of all pre-term deliveries in the United States 
may be attributable to maternal smoking. Third, full-term babies born to 
women who smoke during pregnancy are on the average 200 grams lighter than 
babies born to comparable women who do not smoke. The whole distribution 
of birth weights of smokers' babies is shifted downward, and twice as many 
of these babies weigh less than 2500 grams, compared with babies of 
nonsmokers. This finding has been confirmed by over 45 studies of more 



216 



than half a million births. Birth weight is affected by maternal smoking 
independently and to a uniform extent, regardless of other determinants of 
birth weight. The more the mother smokes, the greater the reduction in 
birth weight of the baby. This dose-response relationship has been 
demonstrated in at least 20 studies. What is not clear from the Surgeon 
General's Report is the circumstances under which maternal smoking 
produces one type of negative outcome {perinatal death, pre-term birth, or 
low birth weight) instead of another. 

Nonetheless, when we juxtapose these findings with reports on the 
prevalence of smoking among American women, we cannot help but speculate 
on some possible hypotheses which may explain the observed U.S. 
white-Chinese differentials in fetal mortality. Might the differentials 
be attributable to the lower prevalence of smoking among Chinese women 
compared with white women? If only data on the prevalence of smoking 
among Chinese women in the United States were available, it would have 
been possible at least to test the plausibility of this hypothesis on an 
aggregate level. Unfortunately, our literature search revealed the 
absence of any recent data base which contains a sample of Chinese 
American women large enough for us to examine this issue. We, therefore, 
need to defer the testing of this hypothesis. 

Smoking, however, is not the only risk factor which might turn out 
to differentiate between white and Chinese women. Alcohol consumption may 
be another factor, although it has not been as thoroughly investigated as 
smoking has been. About the best evidence thus far comes from a 
case-control study on drinking during pregnancy and spontaneous 
abortions. Using maximum- likelihood logistic regression analysis, Klein 
et al. (1977) found that the adjusted odds ratio for this association was 
2.62 — higher than the highest perinatal mortality risk ratio ever reported 
for smoking. With non-drinkers as the comparison group, the adjusted odds 
ratio for drinking daily (2.58) was significantly higher than that for 
drinking twice weekly or more (2.33), indicating a dose-response effect. 
Consideration of wine, beer, and spirits separately suggested that the 
minimum harmful dose was one ounce of absolute alcohol. Several 
potentially confounding variables, including maternal age, gestation, 
prior spontaneous abortions, smoking and nausea or vomiting, were 
controlled for in the analysis. The association between drinking during 
pregnancy and spontaneous abortions did not vary with these factors. Even 
moderate consumption of alcohol during pregnancy is a risk factor for, and 
may be a cause of, spontaneous abortion. Among the possible mechanisms, 
acute fetal poisoning seems the most likely, although chronic poisoning is 
also possible. Klein et al. hypothesized that because their karyotype 
analyses indicate that drinking during pregnancy raised the risk of 
aborting euploid — rather than aneuploid — conceptions, it is quite likely 
that spontaneous abortion is a manifestation of the human reproductive 
sensitivity to alcohol. In this sense, newborns with congenital 
malformations or fetal alcohol syndromes may well be the rare survivors of 
'poisoned' conceptions. 

In view of the evidence provided by these sociomedical researchers, 
we are tempted to posit that perhaps the Chinese American women who gave 
births during 1973-77, being 86 percent foreign-born (non-U. S. residents 



217 



excluded) — compared to about 8 percent foreign-born in the white 
population — may have retained a traditional set of sociocultural health 
habits or lifestyles which contribute to their advantage in fetal 
survival. At least up until very recently, rigid sex-role socialization 
in Taiwan gives men the freedome to smoke and drink but strongly 
discourages women from cultivating such habits. 



IV. Ethnic Differences in General Mortality 

Analysis of the general mortality data reveals that the Chinese 
advantage in survival is not limited only to the perinatal period. Table 
10 shows both crude death rates and age-adjusted death rates. Age 
adjustment, using the direct method, is the application of the 
age-specific death rates in a population of interest to a standardized age 
distribution in order to correct for differences in age structure between 
and among the racial groups being compared. Without the age adjustment, 
differences in the observed rates due to age differences in population 
composition can distort comparisons of overall mortality. Therefore, the 
age-adjusted death rates are what mortality levels would be if age 
distributions were identical for all the racial groups being compared. 



Table 10 



For 1980, the ranking from the highest to the lowest mortality rates 
per 1,000 population are: blacks (8.3), American Indian and Alaskan Native 
(5.8), whites (5.6), Chinese (3.5), and Japanese (2.9). The crude death 
rate of blacks is actually lower than that of the white population, 
demonstrating the importance of age adjustment. 

This ranking of age-adjusted rates for 1980 is identical to that 
found for 1970. Japanese Americans have maintained the lowest mortality 
rates throughout the past 30-year period, while Chinese Americans have 
maintained the second lowest rates for the past 20 years. Data for 
Filipino Americans (not tabulated for 1970) reveal an overall mortality 
rate for 1980 that is even lower than that found for Japanese Americans. 

A body of literature exists (Gordon, 1967; Kitagawa and Hauser, 
1973; NCHS, 1975; King, 1974, 1981) which suggests that socioeconomic 
status influences one's chance of staying alive. However, direct evidence 
of the generalizability of this conclusion to Asian-white differentials is 
impossible to obtain at present because death rates needed to make 
comparisons between these races by socioeconomic status are lacking. 
Indirect evidence is available from the 1980 census which suggests that 
Asians appear to have high socioeconomic status. In aggregate. They are 
proportionately more educated than white or black Americans (see Table 2 
earlier) . Furthermore, notwithstanding the bimodal distribution of 
occupations, more Asians are employed in high-prestige occupations, 
especially in the professional and technical fields. Data on income 
distributions reveal that a higher proportion of Asian families than white 
reported having incomes in 1979 of more than $25,000 (see Table 4 



218 



earlier). At the other extreme, we find that whereas 5.6 percent of the 
white families earned $5,000 or less in 1979, only between 3 to 4 percent 
of the Japanese and Filipinos are in that situation. The figure for 
Chinese (5.9 percent) is comparable to that for white. 

Therefore, by and large, the evidence suggests the presence — on an 
aggregate level — of an inverse relationship between socioeconomic status 
and mortality which may account for some of the observed inter-ethnic 
mortality differentials. Indeed, the following data strongly indicate 
that the effects of socioeconomic status may be operating in age-specific 
deaths, and in mortality from specific causes of death. 



A. Age-specific Differentials 

Examination of the age-specific mortality rates (Table 11) indicates 
that the age pattern of deaths is similar across racial/ethnic groups. At 
the youngest age group (0-5 years) , death rates are high, but they drop to 
a minimum in early childhood (5-14 years) , only to increase steadily to a 
maximum at the oldest ages. One notes, further, that in every age group, 
the rates for Asians are lower than those reported for white Americans. 



Table 11 



B. Race-Mortality Ratios 

By dividing the age-specific death rates for each Asian subgroup 
with the corresponding white rates, we may obtain race-mortality ratios 
for the different ethnic groups. A ratio of 1.0 means that the death rate 
for the minority population is no different from that found for the 
majority white population. On the other hand, a ratio of greater than 1.0 
suggests the number of times by which the minority death rates exceed the 
majority rates. Table 12 shows that for every age group, the mortality 
ratios are less than 1.0 — which signifies that, for All Causes of Death, 
white death rates exceeded Asian rates. 



Table 12 



Sufficient evidence exists to indicate that there is a positive 
association between social class and overall mortality and that the class 
differentials are largest in the middle years of life. Table 12 shows 
that the race-mortality ratios are lowest in the 15-24 and 25-34 age 
groups for Chinese — implying that the differences between white and 
Chinese are greatest during these years. For Japanese, the ratios are 
lowest in the 15-24, 25-34, and 35-44 age groups, increasing slightly in 
the 45-54 age group, but declining markedly in the 55-64 age group — a 
pattern which was replicated by the Filipinos, except for the 45-54 age 
group rise. 



219 



These findings lend confidence to the tentative conclusion that 
perhaps, beyond the issue of errors in filing death certificates, there 
exists a strong socioeconomic factor which accounts for the observed 
white-Asian differentials, especially in the early and middle years of 
life. 



C. Leading Causes of Death 

Heart disease, cancer, cerebrovascular disease, and accidents have 
been the leading causes of death in the United States since around 1950 
(Fingerhut, Wilson, and Feldman, 1980) . Table 13 shows the 10 leading 
causes of death for the United States and the rank order of these causes 
for whites and Asians. Insofar as the first 4 leading causes of death are 
concerned, all four groups have identical rankings. However, they differ 
in proportional mortality. In 1980, heart disease accounted for nearly 
two-fifths (39 percent) of all deaths for white Americans, compared to 
about one-third of the Asian deaths (32 percent for Chinese, 30 percent 
for Japanese, and 34 percent for Filipinos) . On the other hand, cancer 
accounted for a larger proportion of deaths among Chinese (27 percent) and 
Japanese (25 percent) than among whites or Filipinos (each 21 percent) , 
while cerebrovascular diseases made up 11 percent of all deaths for 
Japanese and 10 percent for Filipinos, and about 9 percent for whites and 
Chinese. 



Table 13 



In terms of ranking, we note that pneumonia and influenza rank fifth 
for the Asian groups, compared to sixth rank in the total U.S. population, 
while suicide ranks higher as a cause of death in both the Chinese and 
Japanese populations (seventh and sixth, respectively) than in the 
Filipino group or in the total U.S. population. 

Table 14 shows the age-adjusted race-mortality ratios for specific 
causes of death. Here, we note that in the area of accidents, the 
age-adjusted mortality ratios for Chinese (.34), Japanese (.44), and 
Filipinos (.39) are extremely low. From a public health point of view, 
deaths due to accidents is of considerable interest because it is for this 
particular cause of death that the differential consequences of social 
class will be most visible and detectable. After all, it is only when the 
cause of death is preventable that one's social class is predictive of 
one's ability to command resources to engage in prevention behavior. 
Having said that, we note that for such a preventable cause of death as 
accidents, the Asian mortality rates are only about one-third to two-fifth 
the size of the white majority rate. This finding gives a strong hint of 
the possible combined effects of culture and socioeconomic status on 
deaths due to accidents. 



Table 14 



220 



On the other hand, heart disease, cancer, and cerebrovascular 
disease — the top three leading causes of death in the country and the 
leading ones as well for the ethnic groups under comparison — show a 
different pattern. In general, the ratios for heart disease, where the 
largest number of deaths have been reported, range from 0.4 to 0.5 (if we 
round off the figures) . This finding suggests the possibility that 
perhaps, social, cultural, dietary, as well as environmental factors are 
at play in heart disease — more than we ever realized in the past. The 
near similarities in race-mortality ratios experienced by the three Asian 
groups vis-a-vis white Americans pinpoints new avenues for prevention 
research in heart disease. 

Deaths from atherosclerosis ranks 9th in the white population's 
leading causes of death, but it ranks only 14th for Filipinos, 13 for 
Chinese, and 10th for Japanese. Table 14 indicates that the mortality 
ratios for atherosclerosis and heart disease are fairly similar for 
Chinese (0.57 and 0.54, respectively) and Japanese (0.41 and 0.42, 
respectively) , but not for Filipinos, a group which — for historical and 
cultural reasons — behaves differently from Japanese and Chinese. 

Likewise, low mortality ratios are found for COPD among Chinese, 
Japanese and Filipinos (0.50, 0.34, and 0.31, respectively). It is 
possible that this may be attributable to proportionally more non-smokers 
and/or possibly ex-smokers among these Asian groups compared to whites. 
However, among Asians, behavior such as abstention from smoking may very 
well be confounded with social class. Therefore, to the extent that 
smoking can be a good predictor of social class, to that extent high 
socioeconomic status will be inversely associated with such diseases as 
COFD. 

That lifestyle may be an important factor in the differential 
mortality rates of Asians can be surmised from another cause of death for 
which the age-adjusted mortality ratios are nearly as low as — if not lower 
than — those found for accidents. In Table 14, the ratios for chronic 
liver disease and cirrhosis of the liver, for which heavy alcohol use and 
abuse is known to be a definite risk factor, are striking. Indeed, this 
particular cause of death represents the lowest mortality ratios--that is, 
the greatest Asian-white differential — for Filipinos (0.29) and Japanese 
(0.34, which is identical to that found for COFD), and the second lowest 
ratio observed for Chinese (0.42), next only to accidents (0.34). 

Insofor as cancer is concerned, the age-adjusted mortality 
differences between white and specific Asian groups are small (Table 14) . 
They range from 0.4 to 0.8 depending on which Asian group one looks at. 
Given that cancer is a disease which includes malignant neoplasms of 
various parts of the body, little can be said about the variation in the 
race-mortality ratios for this specific cause of death without a lengthy 
discussion into the genetic vulnerability and socioenvironmental risk 
factors for each type of cancer across the racial groups being compared. 
Moreover, in order to do justice to inter-ethnic comparisons of cancer 
death rates, attention must be given to the proportional mortality of each 
type of cancer in the overall cancer deaths for each group. Chinese, for 
instance, have unusually high morbidity and mortality rates from 



221 



nasopharyngeal cancer (NSP) , a very rare disease. In her review of the 
literature, Yu (1982) found that for all countries of the world in which 
cancer registries exist, the average annual incidence rate for 
nasopharyngeal cancer is less than 1 per 100,000 population (age-adjusted 
to world population) . However, the only areas or populations which have 
reported average annual incidence rates of greater than 5 per 100,000 are: 
San Francisco Bay Area Chinese (19.1 among males; 6.4 for females); 
Singapore Chinese (18.7 for males; 7.1 for females), and Hawaii Chinese 
(10.3 in males; 5.1 in females). These figures excluded data from China 
and Taiwan, where NSP rates are known to be very high but where data were 
unavailable for review at the time. The Chinese proclivity for 
nasopharyngeal cancer may well boost this group's cancer death rate 
vis-a-vis other Asian groups. Research now shows that Chinese have a 
genetic susceptibility for this type of cancer, in addition to a high 
exposure to chemical agents formed from ingestants that popularly consumed 
in the folk diet (Ho, 1979) . 

For cerebrovascular disease, the ratios hover in the range of 0.7 to 
0.8, with Japanese and Chinese showing a ratio of 0.76, and Filipinos, 
0.66. These rates are sufficiently close to suggest greater similarity 
within the Asian group and some amount of differences with white death 
rates. On the other hand, the category of pneumonia and influenza, shows 
as high — if not higher — mortality ratios for two of the Asian groups (0.81 
for Chinese and 0.73 for Japanese). For diabetes mellitus, the mortality 
ratio for Chinese is again relatively high (0.81), but for Japanese, it is 
somewhat lower (0.64), and for Filipinos, even lower (0.49). The reasons 
for these differential ratios by cause of death and across ethnic groups 
are far less clear. Additional review of the literature would be 
necessary to evaluate the relative importance of socioenvironmental risk 
factors for these different diseases and across race. 

One cause of death for which biological factors can be safely ruled 
out as an important one is suicide. Here, we note that the ratios for 
Chinese and Japanese are fairly similar — their rates being three-fifths 
that found for white Americans. The ratio for Filipinos is even lower, 
0.30. These dissimilarities in deaths from suicide within the Asian 
subgroups strong suggest the suicide is closely interrelated with culture, 
besides age and sex. A separate study on suicide between Asians in Asia 
and those in the United States is necessary to search for clues as to what 
accounts most for these observed differentials. 



D. Sex Differences 

In this section, gender comparisons are made within groups and 
between groups. Table 15 shows the within -group differences in mortality 
rates for All Causes of Death. In every ethnic group, the age-adjusted 
male death rate in 1980 was slightly higher than the female rate, 
resulting in a sex-mortality ratio of greater than 1.0. 



Table 15 



222 



Apparently, the most vulnerable age for males in each ethnic group 
is 15-24 years--where the male death rate is two, if not three, times 
greater than the female rate (Table 15) . This is because accident deaths 
take its highest toll in this age range (Table 16) . Among whites, 
accidents account for 61 percent of male deaths in the late teens to early 
twenties, compared to 50 percent for female. Almost comparable sex 
differentials may be observed for Chinese (29 percent for males versus 21 
percent for females) , Japanese (47 percent for males versus 39 percent for 
females) , and Filipinos (54 percent for males versus 43 percent for 
females) . The exceedingly high proportion of accidents among white males 
in this age group is carried into the next age group, mid-20s to early 
30s, where 43 percent of deaths is due to accidents, compared to 26 
percent for white females (Table not shown due to space limitation) . 



Table 16 



Trailing behind accidents as an important cause of death, homicide 
is the second largest major cause of death among men during the vulnerable 
ages between 15 and 24. The proportional mortality is highest for Chinese 
males (26 percent) , followed by Filipinos (15 percent) and whites (13 
percent) , with Japanese the lowest (5 percent) . 

Early adulthood (15 to 24 years) aside, the largest within-group sex 
differentials for Asians may be found in the "young old" — 65 to 74 age 
group — where male death rates surpass female death rates by a ratio of 2.6 
for Filipinos, and about 2.0 for white, Chinese, and Japanese (Table 15) . 
Major cardiovascular diseases and malignant neoplasms begin to take high 
tolls in this age group, regardless of race (Table 17) . Among whites, at 
least one out of every two deaths among the young old of both sexes is 
attributable to major cardiovascular diseases, compared to about 1 out of 
3 (32 percent) for Chinese, just slightly less than that for Japanese (31 
percent) and about 1 out of 5 (22 percent) for Filipinos. Sex differences 
in proportional mortality for major cardiovascular diseases are largest 
for Filipinos (43 percent for males compared with 35 percent for females) 
and Chinese (40 percent for males, compared with 30 percent for females) , 
followed by Japanese (37 percent for males versus 31 percent for females) 
and white Americans (52 percent for males, versus 49 percent for females) . 



Table 17 



Insofar as malignant neoplasms are concerned (see Table 17) , the 

proportional mortality is higher for women than for men in three groups: 

Chinese (36 percent for women versus 30 percent for men) , Filipinos (26 

percent for female versus 22 percent for males) , and white Americans (29 

percent for female versus 27 percent for males) . Among Japanese, no such 
differences were observed; the proportional mortality for men and women is 
31 percent. 



223 



Focusing now on cause- specific within-group sex differences (Table 
18) , we observed that, by and large, for each of the 10 leading causes of 
death, the age-adjusted sex-mortality ratios are greater than 1.0 — meaning 
that males rates again surpassed female rates in each ethnic group. The 
magnitudes of the excess range from only 1.04 for cerebrovascular diseases 
within the Japanese group to as high as 3.81 for chronic obstructive 
pulmonary disease (COPD) in the same ethnic group. Among whites, the 
age-adjusted male death rate surpasses the female death rate by as much as 
3.13 for suicide, 2.93 for accidents, 2.92 for COPD, 2.24 for chronic 
liver disease and cirrhosis, and 2.08 for heart disease. Among Chinese, 
chronic liver disease and cirrhosis are 4.69 times higher for men than for 
women, COPD rates 3.74 times higher; heart disease 2.18 times higher, and 
atherosclerosis, exactly twice as high. The Japanese seem to have similar 
patterns as Chinese, except for COPD and heart disease. For suicide the 
rate for men is greater than that for women, resulting in a sex-mortality 
ratio of 2.22. In contrast, Pilipinos have the largest differences 
between the sexes in many of the leading causes of death. A sex-mortality 
ratio of greater than 2.0 is found for atherosclerosis (3.60), heart 
disease (2.88), suicide (2.70), COPD (2.52), accidents (2.26), pneumonia 
and influenza (2.05), and chronic liver disease and cirrhosis (2.0). 



Table 18 



Searching for between -group gender differences, we compare each of 
the Asian groups with whites, sex for sex (see Appendix) . For All Causes 
of Death, the sex-mortality ratios are consistently less than 1.0. 
However, among the 10 leading causes of death, one may occasionally find a 
few age groups where Asians have a sex-mortality ratios of greater than 
1.0, which signifies that for that particular age-sex group, the Asian 
death rate is greater than the white death rate. Nonetheless, with the 
exception of suicide among Chinese females from middle-age onwards and 
Japanese "old old" females (75 years and over) , the magnitude of excess in 
the Asian over white death rate remains small and defies generalization 
because of its inconsistent age pattern (see Appendix) . For chronic liver 
disease and cirrhosis, for instance, the race-mortality ratios are as high 
as 3,67 and 3.00, respectively for Chinese and the Filipino males less 
than 5 years of age, 3.28 for Chinese females 85 years and over, and 4.70 
for Filipino males in that same age group. Without dismissing the 
possibility of errors in death registration due to random fluctuations, we 
are tempted to hazard a guess that, perhaps, this inconsistencies reflect 
a cohort effect for the oldest group and some kind of genetic 
susceptibility associated with hepatitis or biliary cirrhosis among the 
very young. 



V. The Nativity Factor 

Nativity, or the place where the decedent was born (U.S. or abroad), 
is an important variable in any analysis of Asian mortality, principally 
because, as of the 1980 census, Asians as a whole have by far the largest 
proportion of foreign-born persons (58 percent) of all ethnic groups in 



224 



America. Among Filipinos, the figure is estimated at 66 percent; Chinese, 
at 65 percent; and Japanese 31 percent. Table 19 displays the 
nativity-mortality ratios for All Causes of Death by age. Both the crude 
and age-adjusted nativity-mortality ratios show an excess of death rates 
among the foreign-born vis-a-vis native-born that is larger than the 
within-group sex-differences observed in Table 15. 



Table 19 



No doubt, these differential patterns in mortality by nativity are 
due to the selective waves of Asian immigration which produce different 
cohorts of foreign-born persons in the United States. Each cohort, in 
turn, may have distinctive health habits and mortality patterns. To 
illustrate, the current population of aged Japanese, the Issei , came to 
the United States from the turn of the century up until 1924 when the 
National Origins Act was passed. Their offspring, the Nisei , are 
American-born and American-educated; some may speak Japanese fluently, but 
most others only have a rudimentary command of the language. For Chinese, 
the first-generation immigrants — called Idai — are distinguishable not only 
by their education (or, rather, lack of education), but also by their 
occupations (mostly services, such as laundries and restaurants) , and a 
combination of territorial and ethnolinguistic affiliations (primarily 
non-Mandarin speakers of southern dialects originating from the area 
within a small radius of the mouth of the Pearl River near the city of 
Canton, Guangdong Province in China) . Because of the large influx of 
Chinese immigrants after 1965, another wave of Idai 's are identifiable. 
They tend to be well-educated professionals and businessmen, predominantly 
Mandarin speakers, if not bilingual or multilingual. As for the 
Filipinos, the different cohorts of immigrants are distinguishable not 
only in terms of generation (that is, time of immigration) , but also in 
terms of occupations (Yu, 1980) . The Pensionados and plantation or farm 
workers were among the first to immigrate to the United States, starting 
from about 1903, followed by a substantial number of "Army and Navy" men 
and their families, formed mainly after the Second World War, then came 
the huge influx of professionals — mostly physicians and nurses — after 1965. 

In Table 19, one notes that for certain age groups, the within-group 
nativity-mortality ratios may be no larger than the gender differences 
observed earlier for the same ethnic group (Table 15) but, by and large, 
the nativity differences are consistent and cannot be dismissed as 
insignificant. In Table 20, we begin to appreciate the magnitude of the 
nativity difference. For every one of the 10 leading causes of death, the 
mortality rates for foreign-born exceeded the native born — on average — by 
a ratio of at least 2.0, if not larger. Therefore, in any future study of 
the health of Asians in America, attention must be given to the nativity 
factor in health behavior, morbidity patterns, and mortality outcome. 



Table 20 



225 



VI. Discussion 

Our analyses of the Asian-white mortality differentials from infancy 
to adulthood have failed to yield a substantial amount of excess deaths 
for a majority of the causes of death. Although some excess of Asian over 
white death rates may be found for certain age-sex categories, the fact 
that these excesses were not consistent across age or subgroups is 
worrisome. About the only consistent pattern of excess that one may 
observe from a study of the ten leading causes of death in the United 
States (see Appendix) is the high suicide rate of Chinese women vis-a-vis 
white women. Mortality data for 1979-81 show that the excess begins to 
surface in the 45-54 age group, exactly ten years later than what the 
mortality data for 1969-71 revealed (Table not shown) . This leads us to 
believe that there appears to be a cohort effect in suicide among Chinese 
women which deserves further investigation. Unfortunately, this report 
was restricted solely to the 1979-81 data. 

Our effort did point up significant clues to possible factors behind 
the inter-ethnic differentials in mortality patterns. Using the most 
recent data on Asian American mortality, and based on a review of the most 
current sociomedical literature, we have discovered that Asian mortality 
rates are low for those causes of death for which cigarette smoking and 
alcohol abuse are major contributing risk factors. While reporting or 
classification errors in vital registration may exist, such types of 
errors cannot be entirely dismissed (nor firmly established) . In the case 
of infant mortality, both the literature and the data indicate that such 
errors, even if they exist, do not appear to be sufficient in and of 
themselves to produce the magnitude of the observed differences for 
Chinese versus white. However, available data from at least one 
State--California — do show that, for Japanese, reporting errors are 
sufficiently serious as to inflate, if not misrepresent, the 
Japanese-white differences in infant mortality. 

In the case of adult mortality, we have discovered the importance of 
socioeconomic status, culture, and lifestyles in the health behavior of 
Asians in the U.S. Our confidence in the contributions of these factors 
to mortality is bolstered by several consistent findings: First, the low 
perinatal mortality rates for Asians, Chinese specifically, that has 
persisted for well over a quarter of a century, and the knowledge that 
Chinese women of child-bearing age — especially the foreign born 
population — seldom smoke cigarettes or drink alcoholic beverages, lead us 
to hypothesize that it is most likely the abstention from these substances 
which play a major role in the inter-ethnic differentials in infant 
survival. Second, the low rates observed for other causes of death among 
Asian adults for which the minimal or moderate use of tobacco and alcohol 
have been shown to be major risk factors is consistent with our findings 
on infant mortality and with the knowledge that alcoholism is a relatively 
rare problem in Asian cultures. 

With the exception of the most recent study conducted in Taiwan and 
which has yet to be published, the bulk of research on alcoholism 
indicates that Asians simply do not drink as much nor as frequently as 
Americans. Both biophysiological and cultural factors have been 



226 



implicated in the low prevalence of alcoholism in this ethnic group. 
Wolff (1972) compared Asians (Japanese, Taiwanese, and Koreans) with 
Caucasian adults in terms of alcohol reactions and found that 83 percent 
of the Asian adults showed a marked flushing response and increased 
optical density of the earlobe shortly after alcohol ingestion, compared 
to less than 2 percent of Caucasians who exhibited similar sensitivity. 
In addition, Asian adults experienced more symptoms of discomfort such as 
dizziness, muscle weakness, pounding in the head, and palpitations. These 
physiologic responses were also documented in Asian infants and in 
American-born Asians who were raised on western diets, thereby suggesting 
a strong genetic predisposition for alcohol abstention. 

From a social-anthropological perspective, the concept of "Happy 
Hour" or "cocktail party" is foreign to indigenous Asian cultures; 
Drinking is usually accepted on festive occasions but, then, only if taken 
with other types of food. That one may find drunken Asians at parties may 
not be a reflection of alcoholism but, rather, an indication of their 
inability to tolerate alcohol — what little it takes for them to get drunk 
compared to white Americans. 

From the data presented in the aforementioned pages, it is clear 
that additional studies, focusing systematically on the lifestyle 
differences between Asians and white, would be necessary to test the 
hypotheses concerning smoking and drinking behavior in these populations. 



227 



FOOTNOTE 



*The above conventional infant mortality rate approximates the 
probability of death among infants in a given year. Since the 
numerator and the denominator do not refer to exactly the same cohort, 
possible bias may emerge. The accuracy of the approximation varies 
from one situation to another but depends in general on the annual 
fluctuations in the number of births. This may have some effects on 
the infant mortality rates of small populations with high level of 
immigration, such as the Asian Americans. Therefore, when data are 
available, such as for the case of California, the cohort rate is 
used. The cohort infant mortality rate is based on the linked records 
of births and infant deaths; it describes the true probability of 
infant death. 



228 



Table 1. Asian Population, 1980 and 1970 



United States 



Number 



1980 



1970 



Percent 



1980 



1970 



Total Asian Population 
Chinese 
Pilipino 
Japanese 
Asian Indian 
Korean 
Vietnamese 
Other Asians 

Laotian 

Thai 

Cambodian 

Pakistani 

Indonesian 

Hmong 

All other 



3,466,421 


1,426, 


148 


812,178 


431 


,583 


781,894 


336, 


731 


716,331 


588 


324 


387,223 




NA 


357,393 


69 


510 


245,025 




NA 


166,377 




NA 


47,683 




NA 


45,279 




NA 


16,044 




NA 


15,792 




NA 


9,618 




NA 


5,204 




NA 


26,757 




NA 



100. 
23. 
22. 
20. 
11. 
10. 

7, 

4. 

1. 

1. 

0. 

0. 

0. 



100. 
30, 
23. 
41. 



0.2 
0.8 



iData based on sample. 

2the 1970 data on the Korean population excluded the State of Alaska. 

Source: Bureau of the Census (1983) . 



229 



Table 2. Percent of Population Completing 4 Years or More of College 
By Specified Race, Age and Sex: United States, 1980 





M a 


1 e 


Female 


Race and Age 


Number 


Percent 


Number 


Percent 


Whitel 










25-34 years 


15,400,161 


24.5 


15,394,841 


21.7 


35-44 years 


10,711,364 


25.9 


10,930,907 


15.6 


45-64 years 


18,618,917 


18.2 


20,292,624 


9.6 


65 years and over 


9,210,721 


10.5 


13,730,849 


7.6 


Chinese^ 










25-34 years 


4,453 


57.1 


4,758 


45.6 


35-44 years 


2,601 


55.0 


2,619 


34.4 


45-64 years 


3,742 


30.7 


3,552 


15.0 


65 years and over 


1,391 


18.5 


1,450 


6.8 



Japanese^ 

25-34 years 3,287 49.3 3,517 40.4 

35-44 years 1,939 48.9 2,861 



3,287 


49.3 


1,939 


48.9 


3,878 


23.7 


1,164 


7.9 



Filipino^ 

25-34 years 3,374 33.2 

35-44 years 2,740 47.6 

45-64 years 2,015 31.9 



65 years and ove 1,880 



8.1 



24.7 



45-64 years 3,878 23.7 5,827 8.9 

65 years and over 1,164 7.9 1,442 



4.6 



4,832 


46.3 


3,412 


53.5 


2,911 


27.9 


982 


11.2 



1 Compiled from published census reports. 

2Data are from the 1980 Census Public Use Microdata A (5%) sample. 



230 



c o 

O H 



00 H .H 






o ^ 






a. 

D 
O 



fN CO 

r^ in 



o in 



r^ -" 



o 



en u 
o 

D -Q 



r^ in 

1,0 c^ 



nH o 



VD 00 

m CO 



n -^ 



H 
■n -C 
fa 3 



r^ in 

en o 



■^ en 



CO '^ 



\£> CO 



m o 



o -^ 



o -^ 






o ^ 

in o 



O ui 

e o 



in tj\ 



r^ in 

O CO 



KO ^-^ 



o rn 
n CO 



r- — ^ 






n. 


^ 


h 


n 


O 


-I 


O 


u 


<]) 


CJ 


u 


c 


a> 


4J 


t 






f-; 


en 


u 






*§ v-t 



cl ta 



O rtS H U 

O 1-^ "3 >-t <u cr> 

U 14-1 0) kJ c 



<ll 


n. 




H 


c 


n 


HI 


o 


■r-t 




tn 


01 


«i: 


u« 




(0 



to to 

E U W 



■Hjj O'H cry-i (DO) Qj 



OJ a TD C C U 



■H .-( in u-i aJ 



E -H 1-t OJ u 



01 to 

a 
o 



e c > > u 

10 Ul H o .-» 



.. a 



231 



o 






4) 



o 

D XI 



n rH ^ 



dS 



in 00 



10 



vo in ^^ 

. • CO 





■o 


•D 


(U 


0) 


j: 


4J 


m 


D 


•H 


0,H 




XI 




D 





a 


OJ 


ID 


u 


x: 


0) 


AJ 


S 






e 


w 






XI 
(0 



s 


a, 




e 


c 





to 





•H 




01 


(U 


< 


k4 



4-1 e j:: o 



c -H u) in 



232 



White 


Chinese 


Japanese 


9.7 


5.9 


8.5 


126,098 


258 


332 


10.3 


3.7 


5.5 


133,709 


163 


213 


3.7 


1.7 


1.9 


48,314 


76 


74 


14.1 


5.5 


7.4 


182,023 


239 


287 


12,937,502 


43,747 


38,94: 



Table 5. Fetal, neonatal, postneonatal and infant deaths 
and death rates for specified race: United States, 1973-1977 



Race of Child 



Fetal death rate^ 
Number 

Neonatal death rate^ 
Number 

Postneonatal death rate-^ 
Number 

Infant death rates^ 
Number 

Live births 

Ratio of neonatal to postneonatal 

death rates 2.8 2.2 2.9 



iFetal death rate is defined in this paper as the number of deaths 
prior to the complete expulsion or extraction from its mother of a 
product of conception (which has had at least 20 weeks of gestation) 
per 1000 total births (number of live births and fetal deaths combined) , 

2Neonatal mortality rate is the number of deaths for infants within 
28 days of birth per 1000 live births. 

^Postneonatal mortality rate is the number of deaths for infants 
within 28 days to 365 days of birth per 1000 live births. 

"^Infant mortality rate is the number of deaths for infants under 1 
year of age per 1000 live births. 

Source: National Center for Health Statistics. Figures on fetal 
deaths were calculated by E. Yu from unpublished data. All other 
figures are based on data published annually in the Vital Statistics of 
the United States, 1973-1975, and unpublished working tables on file in 
the Division of Vital Statistics, NCHS. 



233 



Table 5. Fetal, neonatal, and postneonatal mortality rates 

for specified race of mother based on linked birth and death records: 

Single births to California residents, 1973-1977 



Rate per 1,000 births 

Non-Spanish 
Chinese White Ratio 



Fetal death rate 5.5 7.9 .70 

Neonatal mortality rate 6.3 7.4 .85 

Postneonatal mortality rate 3.4 3.5 .97 



Source: Unpublished data from the California State Department of 
Health, calculated by E. Yu. 



234 



Table 7. Single fetal deaths and death rates in California, 
1973-1977, by age of mother and specified race 



Age of Mother 



White 


non- 


-span 




Chinese 




Rate per 1000 




Rate 


per 1000 


Number 


1: 


ive births 


Numbe: 


r live births 


32 




25.95 







NA 


1050 




8.90 


3 




* 


2131 




6.98 


15 




4.64 


2066 




6.97 


40 




5.09 


1077 




8.60 


23 




5.83 


431 




13.7 7 


9 




7.04 


120 




20.87 


2 




* 


7 




23.97 


1 




* 


47 




228.16 







NA 


6961 




7.88 


93 




5.50 



Less than 15 years 
15-19 years 
20-24 years 
25-29 years 
30-34 years 
35-39 years 
40-44 years 
4 5 and over 
Age not reported 
All age groups 



Source: Unpublished data from the California State Department of Health, 
calculated by E. Yu. 



235 



Table 8. Percent distribution of total births 

by age and race of mother. 

Single births to California residents: 1973-1977 



Non-Spanish 
Chinese White 



Less than 20 years 2 14 

20-24 years 19 35 

25-29 years 47 33 

30-34 years 23 14 

35-39 years 8 4 

40 and over 1 1 



Source: Unpublished data from the California State Department of 
Health, calculated by E. Yu. 



236 



Table 9. Fetal death rate by mother's education for specified race: 

United States, 1973-1977 



Mother's education 



Fetal 
deaths 



White 



Live 
births 



FD 

rate 

per 

1000 

total 

births 



Chinese 



Fetal 
deaths 



Live 
births 



FD 

rate 

per 

1000 

total 

births 



Total 122,729 

Not applicable 36,47 8 

0-8 years 5,549 

9-11 years 16,034 

12 years 34,899 

13-15 years 10,427 
16 years and over 7,486 

Not stated 11,856 



13,082,234 
3,421,614 

498,608 
1,842,690 
4,374,402 
1,533,067 
1,202,055 

209,798 



9.29 
10.55 
11.01 
8.63 
7.91 
6.76 
6.19 
53.49 



226 
88 
17 
7 
27 
15 
30 
42 



40,155 
16,706 
2,385 
1,557 
6,619 
3,206 
8,961 
721 



60 
24 
08 
48 
06 
66 
34 



55.05 



Source: Calculated by E. Yu from unpublished data provided by the 
National Center for Health Statistics. 

^Single and multiple births combined. 

2Total births is the sum of live births and fetal deaths. 



237 



Table 10 

Crude and Age-Adjusted Death Rates^, According to Specified Race: 

United States, 1980 and 19702 

19 8 19 7 



Crude Age-Adjusted^ Crude Age-Adjusted^ 
Death Rate Death Rate Death Rate Death Rate 



All Races 8.8 5.9 9.5 7.1 

Black 8.6 8.3 10.0 10.4 

White 9.1 5.6 9.5 6.8 



8.8 


5.9 


8.6 


8.3 


9.1 


5.6 


5.0 


5.8 


3.7 


3.5 


4.0 


2.9 


2.4 


2.5 



9, 


.5 


10, 


.0 


9, 


.5 


7. 


.2 


4. 


.7 


4. 


,2 



American Indian 

or Native Alaskan 5.0 5.8 7.2 8.2 

Chinese-American 3.7 3.5 4.7 4.9 

Japanese-American 4.0 2.9 4.2 3.3 

Filipino American 



1 Excludes deaths by nonresidents of the United States. 

2 Data for "All Races" are published in the NCHS Monthly Vital Statistics 
Report 32, 3 (Supplement), August 11, 1983. Rates for specified race are 
computed by the author from unpublished data. Because some of the nonwhite 
races are extremely small in number compared to white deaths, deaths for the 
3-year period between 1979 and 1981 were averaged. The denominator is 
obtained from published reports of the 1980 Census. 

3 National Center for Health Statistics, published data. 

4 Age-adjusted by the direct method using as the standard population the age 
distribution of the total population of the United States as enumerated in 
1940. Adjustment is based on 10 age groups for 1980 and 11 age groups for 
1970. 



238 



Table 11 

Average Annual Age-specific-'- and Age-adjusted2 Death Rates 

(per 1,000 Resident Population) 

for All Causes of Death 

by Specified Race: United States, 1980^ 



Age 



White 



Chinese 



Japanese 



Filipino 



All ages, crude 



9.1 



3.7 



4.0 



2.4 



Age-adjusted 



5.6 



3.5 



2.9 



2.5 



Under 5 years 


3.0 


1.5 


1.4 


5-14 years 


.3 


.1 


.2 


15-24 years 


1.1 


.4 


.5 


25-34 years 


1.2 


.5 


.5 


35-44 years 


2.0 


1.1 


.9 


45-54 years 


5.4 


2.8 


2.5 


55-64 years 


12.8 


7.1 


5.3 


65-74 years 


29.1 


18.9 


14.5 


75-84 years 


66.2 


52.3 


39.8 


85 years and over 


159.9 


112.8 


131.6 



1.2 

.1 

.4 

.5 

.8 

2.0 

4.0 

13.7 

39.3 

77.6 



^The numerator consists of 1979-81 cumulative number of deaths, 
excluding those of foreign residents; the denominator is based on the 
1980 Census. 

2Age-adjusted by the direct method using the 1940 U.S. population as 
the standard. 

^Source: Unpublished data from the National Center for Health 
Statistics, calculated by the authors. 



239 



Table 12 

Race-Mortality Ratios^ for All Causes of Death 

According to Specified Race: United States, 1980 



Age 


Chinese 


Japanese 


Filipino 


All ages, crude 


0.41 


0.44 


0.26 


Age-ad j usted^ 


0.26 


0.52 


0.44 


Under 


5 years 


0.49 


0.47 


0.40 


5-14 


years 


0.49 


0.56 


0.43 


15-24 


years 


0.33 


0.43 


0.37 


25-34 


years 


0.42 


0.44 


0.37 


35-44 


years 


0.54 


0.44 


0.37 


45-54 


years 


0.51 


0.46 


0.36 


55-64 


years 


0.55 


0.41 


0.31 


65-74 


years 


0.65 


0.50 


0.47 


75-84 


years 


0.79 


0.60 


0.59 


85 years and over 


0.71 


0.82 


0.49 



^Excludes deaths of nonresidents of the United States. Ratios 
are computed by dividing the age-specific death rate of a specified 
ethnic group by the death rate of the white population in that age 
group. 

^Age-adjusted by the direct method, using as the standard 
population the age distribution of the total population of the 
United States in 1940. Adjustment is based on 10 age groups. 

Source: Unpublished data from the National Center for Health 
Statistics, computed by the authors. 



240 



(U 
Q 

M-l 

o 

CO 
0) 

to 

D 
ITJ 

O O 
00 

C rH 

•H 

-o » 
nj w 

(U 0) 
J 4J 
(0 
o -P 
■H CO 



<u 


-n 


x: 


0) 


4J 


-U 




•H 


14-1 


c 





D 



0) 



S 0) 



S >i'a 



(0 


•H 


■u 


O 


14 


(1) 





04 


a 


w 


rH 





(0 


4J 


c 







n> 


•H 


c 



i-i 
CU O 

u 



cu 

c 

(0 

u 
0) 

-o 

I-I 

o 

c 

Di 





• 




S 





• 


c 


Ot 


•H 




a 




■rH 


^ 


H 


c 


•H 


<a 


0< 


« 



Pn 



03 
0) 

c 



0) 


• 


M 


u< 


tt) 




C 




•H 


j^ 


x: 


c 


r) 


<0 




OS 



(0 
Q) 

T! 


U 

I 

Q 
O 



to 

0) 
W 
4J 

-H 

c 

D 



CO 
0) 
U 

D 

m 
u 

c 

•rH 

Xi 

m 



LninrHr^ooooocNvDin 



nOO>J3tN(NrHHOrH 
fO CN rH 



rHCNrO'*i^inr-<3%oao 



■*"*rM-*oi/io<Noro 
• ••••••••• 

otnrHLnrvincNrHi-HCN 

OO <N t-H 



r^cNn'«•aoL^r~o^o\o 



CD'*>X)(N-<*OrH(N(TirM 



rH[~00'*(NfO(Ni-HO(N 



iHcNro'S'voinaoo^or^ 



nro^ocNo'^i^'«i'Lnin 



(Tl-HOOinnCNrHrHHrH 

ro CM 



iHCMrO'S'LnvDr-oocno 



1 

■5)' 
o 



CN 

o 






00 00 CO iTi vo r-- 

cTi o n H (Ti 00 

"J- 1 

I o 

o o 



n CN 
I I 



o 
n 



I I 

o o o 



rH O 



n CO (Ti CO in r~ "^ 



rH^W-^-^cMir)'* 



CO 

0) 

m 



in 

I 

o 
in 

H 



<u 




U} 




(0 


>i 


Q) 


l-i 


W 


D 











5 






•H 




TD 






m 




H 


(0 




01 




Q) 






M 




D 


N 









AJ 






(C 




ft 


£ 




£ 




O 






<u 






Q) 




>-l 




■H 






(0 




(U 


D 




u 




rH 






■H 




> 


iH 




•H 




MH 






D 




•iH 
4J 


H-l 

C 


03 

3 


U 




c 

•H 






u 




U 


M 


-U 


l8 


w 


1 






10 




D 




•H 




■H 


14H 


0) 




r-H 




u 


T3 


rH 


U 


03 


rH 


W 




D 




JJ 


fi 


rH 


0) 





0) 


«3 




u 




M 


10 





> 


U 


U3 


<U 




u 




X! 




s 


■H 


0) 




M 




re 


W 


O 


<0 




J 


rH 


i3 


■H 




> 


■u 




•iH 


w 




U 




Q 







c 


U 


c 


0) 


o 


03 


0) 




u 


1-1 


OJ 


-rl 





4J 


•rl 





TD 


JJ 


d) 


X! 


^3 


c 


e 


0) 


c 


I-I 


•H 


i-i 


U 


0) 


■H 





D 


XI 


o 


0) 


u 


(0 


c 


U 


U 


u 


0) 


(0 


I-I 


x; 


•H 


0) 


ra 


01 


u 


£ 


c 


•H 


x; 


4J 


D 


a 


• 


O 


• 


u 

• 


ft 

• 


D 

• 


• 


< 


W 



rHCNrO'*ini£)|~00(TiO 



03 
IH 
O 

x: 

3 
(0 

0) 

4J 

>( 
£i 

T3 
0) 
4J 

3 

E 
O 

o 



(0 

-a 

(U 

x: 

03 



3 

a 

c 

3 

c 

(0 

-a 

0) 
J3 
03 



3 
ft 



03 

O 
•H 
4-1 

03 
•H 
JJ 

(0 
4J 
M 

X! 

4J 

rH 

(0 
(U 

X 

I-I 

o 

iw 

IH 

(U 

-u 

c 

03 
U 



(0 

c 
o 

•H 
4-) 

m 



0) 

u 

M 

3 

w 



241 



JS 
4J 
10 
(U 

a 
o 

0) 
(0 

D 
(0 

u 



l(-t 

u 
i) 

a< o 
en CO 

U rH 

o 

14-1 - 

to 
m (u 
O -u 

■H (0 
-U -P 

T3 



.H C 
<0 D 

U 
O 

0) 

o 

(0 

■a 

0) 

■u 

(0 

D 
•m 

< 
I 

a> 



o 
c 

•H 

ft 



-H 
(X. 




03 
Q) 

C 



W 

C 

-H 
U 



s: 

(0 

<u 

Q 

o 

03 
<U 
03 
D 
(0 
U 



tN0VD(Tli-(<Ti<J\(ni/10 



oooooooooo 






oooooooooo 



■^ID'Xl'^Oi-lr-ICNt^'^ 

Lnr~r~nLncooo'*in^ 



oooooooooo 











o 




















(0 




















(8 






03 














(13 






•H 














m 






03 




>1 










■H 











1-1 










Q 






x: 




•r-i 










>1 






S-l 




C 










^ 






■H 




•H 










<a 






u 














c 










TD 










o 






TD 




QJ 










e 






c 




4J 






(U 




rH 


(0 




(0 




u 






03 




D 


N 








-rt 






(0 




ftj 


c 




03 




iH 






(U 






<u 




03 




M-l 






03 




<U 


D 




(0 




C 






•H 




> 


.H 




<u 




-H 






Q 




•H 


IM 


03 


03 




1 










4J 


C 


a 


•H 




IW 






U 




O 


M 


jj 


Q 


03 


i-H 






(0 




D 




•H 




■i-i 


03 


0) 




i-H 




S-l 


"O 


.-( 


U 


03 


03 


U3 




D 




JJ 


c 


.H 


0) 


o 




(0 




O 




03 


10 


S 


> 


u 


T! 


0) 




03 




X! 




s 


■H 


0) 


C 


03 




(0 


03 


O 


(0 




J 


.-1 


(0 


-H 




> 


4J 




■H 


03 




u 




Q 




O 


c 


O 


c 


0) 


o 


03 


0) 




M 


1-1 


Q) 


-H 





-u 


•l-l 


O 


■D 


-U 


(U 


Xi 


TD 


c 


B 





c 


1-1 


•H 


u 


o 


03 


-H 





D 


XI 


o 


0) 


O 


(C 


c 


u 


O 


u 


0) 


<o 


1-1 


JZ 


•H 


03 


(0 


03 


o 


x: 


c 


•H 


x: 


4-1 


D 


sa 


u 


u 


< 


u 


ft 


Q 


u 


< 


W 



(0 






03 






XI 






Xl 






03 


• 




4J 


c 




03 


o 




D 


-H 




■n 


■p 




T) 


(0 




(0 


rH 




03 


a 




Oi 







(0 


04 




0) 


03 




x: 


JJ 




-p 


•H 

x: 




171 


s 


03 


c 




O 


■l-l 





•H 


ID 


x: 


+J 


-H 


4J 


01 


> 




•H 


-r^ 


UH 


4J 


■o 





(0 
-P 


>1 


03 


03 


ia 


4J 






10 


x: 


XI 


IH 


4J 


-U 




rH 


(0 


x; 


10 


0) 


jj 





^3 


(0 



ffi 


14-1 


T3 


IH 












TI 


UH 


03 







01 


-P 


U 


D 


03 





(0 


3 


JJ 


u 


•1— 1 


C 




13 





o 


(0 


u 


■^^ 


1 




IW 





rH 


-H 


cr> 


10 


O 


(0 


c 


03 







Oi 





■H 


03 


x: 


4J 




-p 


(0 


x: 




s 


u 


>i 




(0 


XI 





03 




x: 




a 


4J 


l-l 


3 




o 





g 


H-l 


IH 







cn 


IH 


'O 




ItH 


03 


O 




4J 


•H 


(0 


(0 


C 


4J 


rH 


x: 


(0 


3 


-p 


■a 


u 







rH 




TJ 


(0 


TJ 





U 





jz 




•iH 


w 


0) 


MH 


•^^ 


l-l 


•H 


rH 


(0 


o 


X! 







3 


03 


a 


&i 


o 


03 


C 


•H 




D 


JJ 


(0 




10 






« 


MH 


• • 




o 





• • 




U 








u 


4-) 


■p 


3 





(0 





a 


IH 


tn 



242 



Table 15 
Within-Group Sex-Mortality Ratios^ for All Causes of Death: 

United States, 1980 



Age 


White 


Chinese 


Japanese 


Filipino 


All ages, crude 


1.23 


1.63 


1.33 


3.25 


Age-adjusted^ 


1.82 


1.75 


1.65 


1.96 


Under 5 years 


1.28 


1.13 


1.37 


1.01 


5-14 years 


1.56 


0.73 


1.31 


1.39 


15-24 years 


3.00 


2.39 


2.05 


2.80 


25-34 years 


2.61 


1.60 


1.84 


1.87 


35-44 years 


1.86 


1.27 


1.17 


1.38 


45-54 years 


1.88 


1.70 


1.64 


1.34 


55-64 years 


1.98 


1.94 


1.72 


2.18 


65-74 years 


1.96 


1.97 


1.95 


2.58 


75-84 years 


1.65 


1.86 


1.60 


2.07 


85 years and over 


1.28 


1.31 


1.38 


1.71 



^Excludes deaths of nonresidents of the United States. Ratios are 
computed for each ethnic group by dividing the age specific death rate 
of males by the death rates of females in that age group. 

^Age-adjusted by the direct method, using as the standard population 
the age distribution of the total population of the United States in 
1940. Adjustment is based on 10 age groups. 

Source: Unpublished data from the National Center for Health 
Statistics, computed by the authors. 



243 



01 

fa 






,-t 00 
CO •* 



in 



o 

00 



Id 
<u 
JJ 
m 

w 

0) 
4J 
■H 

c 

3 



3 

o 

U 
01 



(0 
(U 

I 
in 






(0 



U3 
J= 
4-1 
(0 
(U 
Q 

•a 

4J 

u 

01 
fH 

0) 
W 

M 
O 

>1 

4J 



(0 
JJ 
U 

o 

S 



a 



O) 



IT3 

e 



0) 
0) 

c 

^-3 



as 

4-1 

O 

Eh 



03 

e 



41 
W 
4) 
C 

•H 

u 



o 

Eh 



e 

0) 

fa 



(0 

2 



•a- 

in* 



CO 



ID 



■>3< 
•3' 






o 
in 



o 



VO ^ 



n o 
n (N 






r-i r^ 



o CO 



'J' 
CN 






O 









o 



00 


r~ CN 


m 


CO 


• 


• • 


• 




o 


\£) ^ 


00 


o 


CN 


H 




CN 






CN o^ 



CN 1-1 rH 



iH 00 



■<3« vo 






CM 



o 

rH 
CN 



CO 



00 
00 



CN 



10 

u 


4J 

3 

nj 
dj 

4-1 

ja 

'O 
01 

4J 

3 

a 

£ 
O 

o 



01 

o 

•-4 
4J 
10 
•H 
4J 
10 
4-1 
M 



(0 

<u 

X 

u 
O 

u 
01 

■p 

c 

0) 
CJ 

l-l 

(0 

c 
o 

■r4 
4-1 
CO 

z 

ai 

x: 

4J 



s: 
S 



o 
a 
o 

u 



o 

Eh 



oo 
in 






W 
(U 
(0 

3 
(0 

u 

T3 
0) 
4-1 

o 

01 

rH 

0) 

w 







4J 


01 






9) 


C 


dJ 






r-l 


<v 


rH 




to 


U 


•V 






4J 


•H 


■H 


T3 




c 


J3 





C 




CD 


01 


o 


ID 




TS 


> 


(0 






•H 






0) 




U 


M 


M 


"U 


0) 


u 





0) 


-H 


T3 


CO 


4J 


j:: 


o 


■r4 







4J 


•rH 


o 


rH 


s 


O 


E 



•H 

3 


< 






K 


w 



CO 

0) 

J3 

to 



XI 

D 

a 

c 

D 



CD 
U 

3 

o 



244 



o 

00 



W 
0) 
-U 

to 

m 

Ti 
<u 
■u 

•H 

c 

D 



O4 
3 

l-l 

o 

<u 
en 

< 

(0 

I 

in 
vo 

<u 

4J 



(0 



x: 
jj 
la 

(U 

Q 

-D 
0) 

4J 

u 
(1) 
1-1 
<u 

CO 



M-l 
>1 



(0 

■u 
u 


S 



(0 

c 
o 

■H 
4J 

o 
o 



a . 



fa 



(0 

a 



<o 



1T3 



10 

s 
fa 



o 

Eh 



10 

e 

01 

fa 



(0 

2 



o 

El 



0) 

in 
3 
nj 
(J 

TJ 
0) 

4J 

O 
0) 
i-l 
01 
W 



LT) 



CM 



o 
n 



rH 



CM 

n 






CM 



rH 



CM 



CM 



CM 



to 

B 

CO 
(0 

l-l 
D4 
o 

c 



c 
c 

CI 



10 

s 






•ST 






o 









o 
n 



o 



1^ 

CO 



CTi 



rH 

in 



o 
in 



CM 

n 






00 

CM 



m 
M 



iH 
ro 



CM 






ra 

rH 

3 

o 

Ul 
(0 

> 

o 



'O 



in 

4) 

in 
10 
a) 
(0 in 

U -H 

u 
O 



•"J 



oo 
n 



CM 



o 

iH 



U 
(0 
01 

x: 

U-l 

o 

in 
01 
u> 
10 
01 
in 



CM 



rH 



CM 






CM 



cn 



00 

CN 



O [- 



in 

CM 



O 






(0 



0) 

in 
10 
01 
in 



M 
10 
0) 

x; 



E 

aj 
x: 
o 
in 



in 
u 


Si 
4J 
3 
(0 

0) 

x: 
jj 

>i 

XI 

^0 
(U 

3 
a 
e 

o 



in 
O 



in 



(0 



w 

x: 



(0 
0) 

ffi 

u 

o 

M-l 



c 

0) 

o 



(0 

c 



-H 
4J 

10 

z 
a> 

e 
o 



03 
4J 
(0 
-O 

"O 
0) 

x: 

CO 



XI 

3 

a 

c 



0) 
O 
u 

3 
O 

cn 



245 



x: 
jj 

(0 
0) 

T3 

IM 


10 
(U 

u 

3 
(0 

o 

c 

-H 

-o 

10 

a) 



O 
4-1 

W 
O 

<0 



0) 
r-i 

XI 
(0 
E-i 



(0 
JJ 
1-1 ■ 


S ; 

X 

(U 

m 
-o 

0) 
4J 

(fl 

3 
■n 

< 
I 

< 

3 

O 

;■! 

oi 

I 

c 

•H 

x: 

4J 



to 

0) 
4J 

m 
J.) 
CO 

0) 



o 

c 



a 



-H 



0) 

w 

0) 

c 

(0 

li 

rO 
1^ 



m 

0) 

c 
u 



■H 

S 



(0 
0) 
Q 

ti-i 
O 

w 

OJ 

in 

3 
(0 

o 

Di 
C 

-H 
T) 
TO 
(U 
►J 



CO 
CO 



^ \D CN un <Tl 

•^r .-N LD o n 



o o o 
o (Ti r~ 



CNiHrHCNJcNCNi-ICMrOCM 



on'*r~i-(oomvD^o(N 



^rHrOr-(Or-(.-(CN) 



ooao(T»oO'^r^oocTto 
r-Hiric— icrir~cocN^£>o 



CN>-li-lrHni-liHM'<MO 



ooaN(TimcNiom'a>nn 
0'*iH<Ti(yir-iHcNni-l 



cNi-iiHrsicNr-lr-HrMi-in 











0) 




















to 




















IS 




















(U 




















(0 






0) 














■H 






in 














Q 






(0 
0) 














>i 






w 




3 










U 






•H 




■n 










>0 






Q 




C 










c 










•H 
















tn 














E 






■H 




T3 






Q) 




rH 


ta 




to 




0) 






w 




3 


N 









JJ 






(0 




CU 


c 




£ 




u 






(U 






0) 




VJ 




■r-l 






u 




0) 


3 




u 




r-l 






-H 




> 


iH 




•H 




M-l 






Q 




-l-l 


C 


3 


U 




C 
■H 






u 




o 


M 


4J 


>j> 


tn 


1 






<o 




3 




•^^ 




■-H 


U-l 


Q) 




r-l 




^ 


^3 


rH 


)-l 


tn 


iH 


tn 




3 




JJ 


C 


iH 


OJ 





tu 


(0 




U 




to 


m 


Q) 


> 


u 


cn 


01 




w 




XI 




S 


■H 


01 




to 




IC 


tn 


o 


nj 




J 


r-( 


oa 


-H 




> 


4J 




-H 


m 




u 




Q 




o 


c 


o 


c 


0) 


u 


m 


01 




u 


u 


(U 


•H 





■u 


•H 





■C 


jJ 


0) 


XI 


X! 


c 


e 


Q) 


c 


1-1 


•i-i 


)-i 


o 


(U 


•H 





3 


XI 





OJ 


u 


(0 


c 


u 


U 


u 


0) 


(C 


u 


x: 


•H 


Q) 


(0 


0) 


u 


x: 


c 


-H 


JZ 


4J 


3 


• 


u 


u 


< 

• 


• 


* 


Q 

• 


u 


< 

* 


• 



r-|CMro'*Lnv£)r~oO(TiO 



x: 


0) 


4J 


x: 


ta 


-U 


0} 




■a 


>i 




^ 


xi 




0) 


■o 


-U 


cu 


tn 


■u 


3 


3 


•r-l 


a 


73 


a 


<0 





1 


u 


OJ 




Oi • 


(0 


ITJ Oi 


-U 


3 


(0 


(U o 


XJ 


JZ VJ 




JJ CTi 


■n 




a; 


Di U 


X! 


C -H 


CO 


■H C 


■H 


T) x: 


H 


■H JJ 


ja 


> d) 


3 


•H 


04 


TD 0) 


c 




3 


>i M 




X! tn 


'O 




c 


x: 0) 


(tS 


4J x; 




(0 -U 


T3 


0) 





-a c 


x: 


•H 


in 


IH 


-rH 


tn 


<-{ 


0) 


Si 


0) rH 


3 


tn (0 


Oi 


3 e 




to Q) 


tn 


u y-i 


u 




•H 


O M 


JJ 


•H O 


tn 


n-i y-i 


•H 


-H 


-u 


U (U 


(0 


(U iJ 


JJ 


a (0 


en 


CO U 






x: 


x: TJ 


jj 


U Q) 


rH 


(0 JJ 


(tJ 


tu in 


(U 


3 


K 


1-1 -r-l 




tj 


1-1 


U-l (C 





1 


iw 


TD Q) 




0) Oi 


1-1 


JJ <tj 


<U 


3 


jJ 


a 0) 


C 


e J2 


d) 


JJ 


U 


o 




>1 


iH 


0) J3 


(0 


l-i 


C 


(tj in 


o 


<u 


■H 


W rH 


JJ 


O (tJ 


(0 




s 


JJ 




(C 1-1 


• 


ffi o 


•• CO 


y-i 


(U l-l 


• • 


CJ 


d) 0) 


u x; 


JJ JJ 


3 JJ 


Id 


3 


3 M 


CO (0 



246 



Table 19 
Nativity-Mortalityl Ratios2 for All Causes of Death: 

United States, 1980 



Age 


Chinese 


Japanese 


Filipino 


All ages, crude 


3.58 


4.78 


5.73 


Age-ad j usted^ 


2.30 


3.04 


2.34 


Under 5 years 


2.52 


2.46 


4.13 


5-14 years 


2.27 


5.03 


3.68 


15-24 years 


2.82 


2.86 


1.34 


25-34 years 


1.52 


1.63 


1.27 


35-44 years 


1.69 


1.59 


0.93 


45-54 years 


1.50 


1.91 


1.18 


55-64 years 


1.68 


4.78 


0.89 


65-74 years 


5.78 


3.69 


2.86 


75 years + 


1.88 


2.52 


16.83 



lExcludes deaths of nonresidents of the United States. 

2Ratios are computed by dividing the age specific death 
rate of foreign born by the death rate of the native born 
population in that age group. 

3Age-adjusted by the direct method, using as the standard 
population the age distribution of the total population of 
the United States in 1940. Adjustment is based on 9 age 
groups. 

Source: Unpublished data from the National Center for Health 
Statistics, computed by the authors. 



247 



























(1) 




(U 



























X! 




x: 




c 






















■U 




JJ 




-H 






















■O 








a 


r- 


ro 


1,0 


<T\ 


in 


CM 


CN 


VO 


m 


in 


0) e 




>i 




•H 


'^ 


^ 


t~ 


CN 


r~ 


<n 


1" 


o 


<Tl 


r-i 


JJ 




X) 




iH 


• 


• 




• 


• 


• 


• 


• 


• 


• 


CO u 








•H 


(M 


^ 


CN 


CS 


rH 


CN 


CN 


CM 


O 


rH 


3 ItH 




T3 




h 






















■n 

-a c 

lO 




0) 
JJ 

3 


•• 
























1 -H 




a 


JS 
























QJ JJ 




s 


4J 
























01 (0 







(0 
























(0 rH 




o 


0) 


0) 






















3 






Q 
























0) a 
x: 




(0 

JJ 


>4-l 


C 


>* 


CN 


iH 


>* 


OO 


r~- 


cn 


in 


CO 


a^ 


JJ Qj 




(0 





10 

ft 


O 


00 

• 


CN 


(Tl 


CO 


• 


o 

• 


• 


CO 

• 


ID 

• 


cn c 




-a 


(0 


m 


ro 


(N 


"a- 


CM 


CN 


«* 


CM 


■^ 


VO 


CM 


c u 




-a 


<u 


l-D 






















-H 




(U 


CO 
























T3 ja 




X! 


. 3 
























•H 




CO 


(0 
























> (U 




•H 


u 
























-H > 




rH 

XJ 


CT" 
























JJ 




3 


£ 
























>i fO 




CU 


-H 
























X! C 




c 


-o 





























3 


Id 


M 






















^ O 






(U 


Q) 


CO 


rH 


n 


n 


rH 


a\ 


CN 


o 


in 


rH 


x; x: 




^3 


1-^ 


C 

■H 


<n 


• 


<yi 

• 


• 


n 

• 


CN 

• 


in 

• 


in 

• 


• 


• 


JJ JJ 




C 
(0 


00 

"^ 


£ 


iH 


CM 


CM 


CM 


ro 


CN 


rH 


CM 


CN 


CM 


0) «H 






u 






















n 




T3 

x: 
























JJ 




cn 


in . 
























<o 




■H 


able 20 
Ratios 
States, 
























0) 1-1 




H 
























CQ 




XI 
























3 Tl 




3 
























10 0) 




ft 
























O JJ 




























cn 




^ 
























U 3 




CO 


























■H -n 




o 
























MH TD 




•rH 
























-H (0 




JJ 
























U 1 




CO 


4J ^ 












(1) 












d) 0) 




■rH 












01 












04 CT1 




JJ 












nj 












CO (0 




(0 


U 












<U 
















JJ 















[fl 






01 






x: 0) 




cn 


f 












■H 

Q 






10 




>1 


u x: 

10 JJ 




X! 


>1 


















(U 




1-1 


Q) 




jJ 


4J 












>t 






CQ 




3 


>. 




^ 


-H 












u 






•H 




■n 


u XJ 




(0 


> 












<a 






Q 




C 


O 




0) 


■H 


x: 










c 










■H 


MH C 




a 


+1 


4J 
















CO 






l-< 






B5 


(0 










g 






•H 




TD 


^ 




u 


Z 


0) 






0) 




i-H 


10 




en 




Q) 


TD J2 









a 






(0 




3 


N 









JJ 


0) 1 




14H 


■o 








m 




PU 


c 




x; 




U 


JJ c 






0) 


M-l 






0) 






Q) 




1-1 




•H 


(0 C71 




U 


+) 









m 




0) 


3 




u 




rH 


rH -H 




d) 


CO 








•H 




> 


rH 




-H 




«H 


3 d) 




JJ 


3 


W 






Q 




-H 


MH 


w 


u 




c 


U M 




c 


•n 


0) 










4J 


c 


3 






•rl 


rH 




(U 


"3 


to 






VJ 




o 


l-l 


JJ 


i^a 


cn 


1 


to 14H 




o 


«C 


3 






(0 




3 




•H 




•H 


IM 


o 


• 




1 


(0 


<u 




iH 




1-1 


T3 


H 


1-1 


CO 


i-{ 


0) 


a 


H 


V 


u 


CO 




3 




JJ 


C 


rH 


Q) 





0) 


(u x; 


3 


(0 


01 




10 




O 




tn 


(0 


0) 


> 


1-1 


cn 


U JJ 





c 


< 


CP 


0) 




M 




XI 




s 


•H 


01 




(0 


1-1 







c 


w 




<0 


W 


O 


10 




J 


rH 


^ 


MH 


C71 


■H 




-H 


•H 




> 


4J 




•H 


w 




U 




ui 




JJ 




-a 


Q 







c 


o 


c 


OJ 


u 


CO 


0) 





CJ 


(0 




(0 




U 


M 


0) 


•H 





4J 


•H 





T3 


•H 0) 


■H 


z 




(U 


JJ 


(U 


XI 


'O 


c 


£ 


0) 


c 


1-1 


-r( 


JJ JJ 


c 






tj 


V4 


U 


0) 


-H 





3 


XI 





0) 


u 


(0 (0 


x: 


• 






m 


c 


1-1 


u 


1-1 


0) 


10 


1-1 


XI 


•H 


cc; u 


JJ 


•• cn 




O 


0) 


m 


OJ 


u 


x: 


c 


■H 


£ 


4-1 


3 




(U 


0) U 




■H 


• 


• 


u 


< 


u 

• 


cu 


Q 


o 


< 

• 


cn 

• 


•• x: 

0) JJ 
JJ (0 




U 

u x; 

3 JJ 






rH 


CN 


n 


■sr 


in 


VD 


t~ 


CO 


a\ 


o 

rH 


<D 


(0 
CO 


3 

cn 10 



248 



REFERENCES 



Breslow, L. and B. Klein 

1971 "Health and Race in California. 
61, 4 (April): 763-75. 



American Journal of Public Health 



Doi, Mary L. , Chien Lin, and Indu Vohra-Sahu 

1981 Pacific/Asian American Research; An Annotated Bibliography . No. 1, 
Bibliography Series. Chicago: Pacific/Asian American Mental Health 
Research Center. 



Fingerhut, L. A., R. W. Wilson, and J. J. Feldman 

1980 "Health and Disease in the United States. 
Health 1: 1-36. 



Annual Review of Public 



Frost, F. and K. Shy 

1980 "Racial differences between linked birth and infant death records 
in Washington State." American Journal of Public Health 70: 974. 



Gordon, T. 
1967 



"Further mortality experience among Japanese-Americans. 
Health Reports 82: 973-984. 



Public 



Ho, J. H. 
1979 



"Some epidemiologic observations on cancer in Hong Kong." National 
Cancer Institute Monograph 53 (Nov) , Second Symposium on 
Epidemiology and Cancer Registries in the Pacific Basin, NIH 
Publication no. 79-1864: 35-47. 



Kan, Stephen H. and William T. Liu 

1984 "The poverty class of Asian Americans." Paper presented at the 

Eastern Sociological Society Annual Meetings in Boston, March 8-10. 

King, Haitung 

1974 "Selected epidemiologic aspects of major diseases and causes of 

death among Chinese in the United States and Asia." Pp. 487-550 in 
Arthur Kleinman et al. (eds.). Medicine in Chinese Cultures . U.S. 
Department of Health, Education, and Welfare Public Health 
Service. National Institutes of Health DHEW Publication No, (NIH) 
75-653. 

King, Haitung and Frances B. Locke 

1981 "Chinese in the United States: A Century of Occupational 

Transition." International Migration Review 14, 1: 15-42. 

Kitagawa, Evelyn M. and Philip M. Hauser 

1973 "Differential mortality in the United States: A study of 

socioeconomic epidemiology." Cambridge: Harvard University Press. 



249 



Kline, J. et al. , 

1977 "Drinking during pregnancy and spontaneous abortion." Lancet 176. 

McCarthy, B. J. et al., 

1980 "The underregistration of neonatal deaths: Georgia, 1974-77." 
American Journal of Public Health 70: 977 

National Center for Health Statistics 

1980 Health, United States 1979 . Washington, D. C. : U. S. Government 
Printing Office. 

National Center for Health Statistics 

1975 Selected Vital Statistics in Poverty and Nonpoverty Areas of 19 

Cities, United States, 1969-71. By S. Ventura, S. Taffel, and E. 
Spratley, Vital and Health Statistics . Series 21-No. 26. DHEW 
Publication No. (HRA) 76-1904. Health Resources Administration. 
Washington, D. C. : U. S. Government Printing Office. 

Norris, F. D. and P. W. A. Shipley 

1971 "A closer look at race differentials in California's infant 
mortality 1965-67. HSMHA Health Reports 86: 810. 

Office of Smoking and Health 

1979 Smoking and Health: A Report of the Surgeon General . Government 
Printing Office. Washington, D. C. 

Office of Smoking and Health 

1980 The Health Consequences of Smoking on Women. A Report of the 
Surgeon General . Washington, D. C: Government Printing Office. 

Patterson, J. E. 

1980 "Assessing the quality of vital statistics." American Journal of 
Public Health 70: 944. 

Vohra-Sahu, Indu 

1983 The Pacific/Asian Americans: A selected and annotated bibliography 
of recent materials . Chicago: Pacific/Asian American Mental Health 
Research Center. 

Wolff, P. H. 

1973 "Vasomotor sensitivity to alcohol in diverse Mongoloid 

population." American Journal of Human Genetics 25: 193. 

Yu, Elena S. H. 

1980 "Filipino migration and community organizations in the United 
States." California Sociologist 3, 2 (Summer): 76-102. 

Yu, Elena S. H. 

1982 "The epidemiology of nasopharyngeal cancer in Chinese 

populations." Unpublished paper, Division of Epidemiology, 
Columbia University. 



250 



Yu, Elena 

1982 "The low mortality rates of Chinese infants: Some plausible 

explanatory factors." Social Science and Medicine 16: 253-265. 

Yu, Elena S. H. , Alice K. Murata, and Chien Lin 

1982 Bibliography of Pacific/Asian American Materials in the Library of 
Congress . Chicago: Pacific/Asian American Mental Health Research 
Center. 



251 



Physical and Mental Health 
Status Indicators for 
Asian-American 
Communities 




Elena S. H. Yu, Ph.D. 

Associate Professor of Sociology in Psychiatry 

and Research Associate 

William T. Liu, Ph.D. 

Professor of Sociology and Director 

Paul Kurzeja, M.A. 
Graduate Research Assistant 

with the assistance of 
Phyllis Flattery 

Pacific/Asian American Mental Health Research Center 
University of Illinois at Chicago 
Chicago, lUinois 



ACKN0WLEDG^4ENT 



The authors are grateful to Thomas F. Drury for his guidance and 
assistance during an earlier study conducted in collaboration with 
Elena Yu and William T. Liu, portions of which are summarized in this 
report. In addition, the critical comments of Mary Doi who read an 
earlier draft of this report are very much appreciated. 



254 



PHYSICAL AND MENTAL HEALTH STATUS INDICATORS FOR 
ASIAN/PACIFIC AMERICANS 



In the last decade, the number of Asians and Pacific Islanders in the 
United States more than doubled, increasing from 1.5 million in 1970 to 3.5 
million in 1980 (Bureau of the Census, 1981) . This dramatic growth came about 
largely through legal immigration made possible by the 1965 Immigration Act 
(Keely, 1971) , which took effect in 1968, and the 1975 Refugee Act following 
the fall of South Vietnam and other parts of IndoChina. The surge in the 
Asian/Pacific Islander population during the 1970s exceeded that for all other 
ethnic groups including persons of Spanish origin. If the present trend 
continues, an even greater increase in the Asian/Pacific American population 
is expected by the 1990s. Yet, despite increased understanding of minority 
health concerns, the health status of minority persons of Asian and Pacific 
Island backgrounds is perhaps the least understood (Weaver, 1976; Lieu et 
al. , 1976; True, 1980). A major reason for this has been the unavailability 
of the necessary statistics for this population. Special studies, 
particularly at the local level, would be a major contribution in filling this 
gap. But, lacking such studies and in a time of scarce resources, alternative 
procedures for developing information should be explored. One such approach 
is secondary analysis of unpublished data available from the National Center 
for Health Statistics (NCHS) . 

As the federal agency mandated to collect, analyze, and disseminate 
national health statistics and epidemiologic data, the National Center for 
Health Statistics maintains three independent sets of records which provide a 
modicum of information on the health of Asian/Pacific Islanders: The National 
Health Interview Survey (NHIS) , the National Ambulatory Medical Care Survey 
(NAMCS) , and the birth and death files for the entire United States. With the 
exception of an article by Yu, Drury, and Liu (1981) , and another paper by Yu 
and Cypress (1982) , data on Asian/Pacific Americans based on the first two 
resources have, for the most part, remained unpublished. On the other hand, 
data on Asian/Pacific Americans from the birth and death files related to 
infant mortality have been analyzed and published (Yu, 1982) , while data on 
adult mortality experiences, with specific emphasis on race, sex, age, and 
nativity differences, have been presented in a companion piece (Yu et al., 
1984) . 



I. Objectives 

The purpose of this paper is to provide some indicators of the physical 
and mental health status of Asian/Pacific Americans. We will review the 
unpublished data at NCHS in order to provide some indicators of their physical 
health status. Data on the mental health of Asian/Pacific American are 
unavailable on a national basis, and small sample, local-area studies are few 
and far between. Nonetheless, a few available studies, some of which are 
still in progress, will be reviewed to provide some indicators of the mental 
health status of this small yet growing minority group. 



255 



II. National Health Interview Survey (NHIS) Data 

The National Health Interview Survey, which has been conducted since July 
1957, consists of a continuous sampling and interviewing of households 
nationwide. Although some changes in its design were made in 1959, 1963, and 
1973 (NCHS, 1975) , a basic concept has persisted throughout. Each week a 
probability sample of households in the civilian noninstitutionalized 
population of the United States is interviewed by personnel of the U.S. Bureau 
of the Census. Consolidation of samples over a time period, e.g., a given 
year, produces estimates of average characteristics of the U.S. population for 
that year. Since the late 1970s, the usual NHIS annual sample has consisted 
of approximately 41,000 eligible households, which yields a probability sample 
of about 120,000 persons (NCHS, 1978). Based on verbal responses to an 
interview schedule, information is collected on the health and other 
characteristics of each member of the household. 

Until 1976, persons of Asian/Pacific Islander backgrounds were identified 
in the NHIS as persons of "other race" based on interviewer observations 
(Drury, Moy, and Poe, 1980) . Since then, as a byproduct of efforts to improve 
the quality of the racial classification used in the sample, respondents of 
Asian or Pacific Islander backgrounds have been allowed to identify themselves 
in the interview as "Asian or Pacific Islander." Despite this improvement, 
the NHIS in any particular year cannot be expected to draw a large enough 
sample to produce estimates for persons of Asian/Pacific Islander backgrounds. 

Certain kinds of health information, referred to as "core" items, are 
collected annually by the NHIS. These include items on the incidence of acute 
conditions, disability days, limitations of activity, perceived health status, 
prevalence of selected chronic conditions, and use of physicians, dentists, 
and of short-stay hospitals. For each of these types of information, it is 
possible to produce tabulations for Asian/Pacific Americans based on multiple 
years, but not single years. This is because Asian/Pacific Americans, 
representing no more than 1.5 percent of the population counted in the 1980 
census are, statistically speaking, "rare elements" in the sampling universe 
of the U.S. population. Any national sample drawn on a probability 
proportional to population size basis, as this one is, will yield too few 
cases of Asians and Pacific Islanders to provide statistically reliable 
estimates. In fact, only about 1,557 NHIS sample persons have been identified 
as Asian/Pacific Americans in an average year since 1976. However, pooling 
these data yields a subsample size of 3,009 sample cases for the two-year 
period 1976-77; 4,465 cases for the three-year period 1976-78; and 6,228 cases 
for the four-year period 1976-79. 

Other kinds of health information, referred to as "supplementary" items 
or supplements, are collected conjointly with the core items, either 
periodically or on a one-time basis. Between 1976 and 1979, supplementary 
information collected included: diabetes (1976) , health practices (1976 and 
1977) , health insurance coverage (1976 and 1978) , problems getting medical 
care (1977) , needs of the disabled for special services (1977) , 
characteristics of usual places of care and blood donor ship (1978) , and home 
care due to a disability or other health problem (197 9) . Much of this 
information has been collected for the total sample, although on some topics 
these data were elicited only from a one-third subsample of persons 20 years 



256 



and older. Still other information is available only for persons who met 
certain screening criteria in the interview. 

In what follows, we will examine the most recent NHIS pooled data for 
Asian/Pacific Americans on five commonly-used health status indicators, and 
report on some other unpublished NHIS data from single years. Whenever 
possible, comparable indicators available from published sources for the 
United States as a whole or for white Americans are noted, even though 
complete tables are not presented here. 



A. Estimates of Selected Health Characteristics 

Table 1 shows the average annual estimates of selected health 
characteristics of Asian/Pacific Americans by age and sex for the period 
1976-78. It further gives the standard errors for these estimates and the 
sample sizes upon which these estimates are based. 

The first column shows the proportion of the population with limitation 
of activity, which is a measure of the long-term impact of chronic disease. 
This measure includes people who, because of such diseases, are unable to 
perform their usual activity (such as working, keeping house, or going to 
school) , or who are limited in the kind or amount of that activity, or are 
limited in other activities. The 1976-78 average annual estimate of 7.0 
percent for Asian/Pacific Americans is considerably lower than the 13.5 
percent for the entire U.S. population in 1977 (NCHS, 1978). Lumping NHIS 
data over the three years 1976-78 proves useful in that the standard errors 
for these estimates are reduced. In no age group is the standard error more 
than 4.3 percent. 

There are marked differences between age groups in regard to limitation 
of activity, with the elderly having the highest rate (36.3 percent). This 
rate for Asian/Pacific Americans, however, is still lower than the 43.0 
percent reported for all persons 65 years or older in the U.S. In all four 
age groups described, the proportion with activity limitation is slightly 
larger for males than for females (2.3 percent compared to 1.4, 5.6 to 4.2, 
15.7 to 12.0, and 39.5 to 32.5, respectively). Such a pattern is consistent 
with that found for the country as a whole, although in every age group, the 
rates for Asian/Pacific Americans are lower than those for the total United 
States. 

Looking now at the second column of Table 1, only about 9 percent of the 
Asian/Pacific American sample perceived their health as either "fair" or 
"poor" in response to the question: "Compared to other persons your age, would 
you say your health is excellent, good, fair, or poor?" This figure is 
relatively low compared to about 12 percent annually for Americans generally 
from 1976 to 1981. Sex differences among Asian/Pacific Americans in 
self-perception of health are minimal, both overall and across age groups. 

Two other measures displayed in Table 1 (physician and dental visits) 
give some indication of the frequency of visits made by Asian/Pacific 
Americans to specific health practitioners. A physician visit is defined as 
consultation with a physician, in person or by telephone, for examination. 



257 



diagnosis, treatment, or advice. Such visits include services provided by a 
nurse or other person acting under a physician's supervision. For the purpose 
of this definition "physician" includes doctors of medicine and osteopathic 
physicians. 

Like other Americans, more Asian/Pacific Americans made a visit to a 
physician's office in the past year (73 percent) than to a dentist's (48.2 
percent). The corresponding rates for the country as a whole are 75.1 percent 
for physician visit and 49.8 percent for dental visit. In general, men made 
fewer visits than women. Among Asian/Pacific Americans, women across age 
groups reported uniformly high proportions of physician visits within the past 
year, while the distribution for males by age is less consistent. Nearly 
equal percentages of men under 17 years of age and those between 45-65 years 
(76.5 percent and 73.6 percent, respectively) visited a physician in the past 
year, compared to a lower rate (62.1 percent) of those in the 17-44 age group 
(62.1 percent) and a higher rate (81.3 percent) of those in the highest age 
group. This pattern differs from that found for the U.S. (see NCHS, 1978: 31). 

A dental visit is defined as any visit to a dentist's office for 
treatment or advice, including services by a technician or hygienist acting 
under a dentist's supervision. Table 1 indicates that while physician visits 
made by Asian/Pacific Americans seem to increase with age (except for men in 
the 17-44 age group) , dental visits show just the opposite pattern, with the 
older age groups making proportionally fewer visits than the younger ones. 
This pattern occurs for men as well as for women, and is consistent with 
findings for the total U.S. 

Next, we examine the proportion of Asian/Pacific Americans who reported 
having one or more short-stay hospital episodes in the past year. Short-stay 
hospitals are: general hospitals; maternity hospitals; eye, ear, nose, and 
throat hospitals; children's hospitals; osteopathic hospitals; and the 
hospital departments of institutions. Overall, the proportion of 
Asian/Pacific Americans who had such hospital episodes increases with age. A 
breakdown by sex, however, reveals some divergent age patterns. 

In the 17-44 years age group, the larger proportion (12.4 percent) of 
women compared to men (4.3 percent) who had one or more short-stay hospital 
episodes in the past year is most likely attributable to maternity services. 
However, in the oldest age group, there is a large disparity between the 
proportion of short-stay hospital episodes for men (17.5 percent) compared to 
women (5.4 percent). We suspect that the inability to care for oneself or to 
be taken care of by others (because of living alone, being without social 
support, or having a complicated chronic illness) , and the ability to pay (as 
reflected in income, retirement benefits, or medical insurance coverage) are 
the critical explanatory factors of this difference. 



B. Data on Health Practices of Asian/Pacific Americans (1977) 

Data on the health practices of Asian/Pacific Americans (Table 2) was 
obtained on a one-third subsample of persons 20 years and over in the 1977 
National Health Interview Survey. Close to 57 percent of Asian/Pacific 
Americans ate breakfast everyday, 21 percent of them sometimes did, and only 

258 



22 percent of them rarely or never did. One-third of those interviewed 
snacked everyday, 38 percent snacked sometimes, and 29 percent rarely or never 
snacked. Only one out of five persons reported needing 6 hours of sleep or 
less, seven out of ten persons usually slept 7-8 hours daily, and another one 
out of ten slept longer hours. About half believed they were as active as 
others, and another one-third believed they were more active than others. 

Insofar as smoking is concerned, 61 percent of Asian/Pacific Americans 
had never smoked, another 11 percent were ex-smokers, and close to three out 
of ten (or 28.7 percent) were smokers. Some 32 percent of Asian/Pacific 
Americans had never used alcohol, while 52 percent reported taking alcoholic 
beverages sometimes. The proportion of persons who drank alcohol once or 
twice a week (8.6 percent) is quite similar to that for persons who drank 
three or more times a week. Because of the small sample size, further 
analysis of the data by sex or other variables is not advisable. 



C. Characteristics of Sources of Health Care 

In the 1978 National Health Interview Survey, data were collected on the 
characteristics of the regular sources of health care for Asian/Pacific 
Americans (Table 3) . Such information was obtained only from 81 percent of 
the total sample of 1,456 persons. Of those who had a regular source of care, 
the place of care in 83 percent of the cases was in a doctor's office, and in 
about 13 percent of the cases, a hospital clinic, emergency room or health 
center. 

Three-quarters of those with a regular source of care had one particular 
physician whom they consulted for medical problems. For some 19 percent of 
those with a regular source, the time required to travel from their home to 
that source was only 10 minutes. For some 38 percent, it was between 10 and 
20 minutes. About 27 percent required a travel time of 30 minutes or longer. 

For those who did not have a regular source of care, the most common 
reason given was that they "haven't needed a doctor" (55 percent). One out of 
eight (12.3 percent) tended to see different doctors for different reasons 
and, hence, did not have a regular source of care. A similar proportion of 
persons attributed the lack of such care to the fact that they had just 
changed their residence. About one out of nine reported that they had not 
found the right doctor or their former doctor was unavailable. 



D. Problems of Meaning and Measurement in NHIS Data 

The NHIS data for Asian/Pacific Americans have three types of 
concept-measurement problems which limit their usefulness: (1) 
conceptualization and empirical identification of this population; (2) the 
conceptualization and measurement of their health characteristics in the 
context of an interview survey; and (3) interpretation of the correlates of 
health characteristics. 



259 



(1) Problems in Conceptualization and Identification of Asians 

From the standpoint of secondary data analysis, perhaps the most basic 
conceptual point to realize is that existing NHIS data cannot be disaggregated 
to identify the diversity of peoples included under the rubric Asian/Pacific 
American. Unlike Hispanic Americans who share a common linguistic root, 
Asian/Pacific Islanders do not share a common language or descent, either in 
the U. S. or abroad. This heterogeneity is amply illustrated by the fact that 
the 1980 Census enumeration codes for Asian Americans and Pacific Islanders 
included no less than 20 categories. For most purposes, the term designates 
residents of the United States from the following countries and territories: 

East Asia ; China (including Taiwan and Hong Kong) , Japan, and Korea. 

Southeast Asia ; Philippines, Vietnam, Cambodia, Laos, Thailand, 
Malaysia, Singapore, and Indonesia. 

Indian Subcontinent or South Asia ; India, Pakistan, Bangladesh, 
Sri Lanka, and Burma. 

Pacific Islands ; Hawaii, Guam, Samoa, Tonga, Fiji, and other Micronesia 
Islands. 



In practice, however, only a few of these groups are identified in federal 
health and other information systems; Chinese, Japanese, Filipinos, and 
Vietnamese or IndoChinese. Despite the fact that a wide spectrum of 
linguistic, cultural and racial diversity exists even among these four major 
groups, the practice of lumping them together into a single ethnic group is 
accepted in many federal, state, and local programs. 

With respect to sample selection, since the probability of selection is 
proportional to population size, persons of Asian/Pacific Islander backgrounds 
are presumably included in the sample in proportion to their representation in 
the civilian, noninstitutionalized population. The composition of the NHIS 
sample of Asian/Pacific Americans most probably reflects, therefore, the 
numerical frequency of particular Asian/Pacific American subgroups within the 
population as counted in the 1970 Census. 

With respect to sample coverage, despite the 96 percent household 
response rate in the NHIS, it should be recognized that health problems among 
Asian/Pacific Americans may be understated by the available data. Ill health 
and other health problems may be unreported or not reported at all because of 
language and other cultural barriers or unwillingness to participate in the 
NHIS. 



(2) Problems in the Conceptualization and Measurement of Health 

A similar point can be made with respect to the measurement of health 
characteristics per se . Information obtained through interview surveys is 
subject to a variety of known limitations (Feldman, 1960; Finkner and 
Nisselson, 1978; Suchman, 1967) . Whether these standard problems have unique 
features in surveys of persons of Asian/Pacific American backgrounds, however. 



260 



is not known. It has been suggested, for example, that health interview 
surveys of minority subpopulations are likely to present unique problems due 
to cultural biases in the wording of questions, inaccuracy of proxy responses 
in sensitive topic areas, and other response errors because of unfavorable 
interviewer-respondent interactions (Salber and Beza, 1980; Rice, Drury and 
Mugge, 1980) . But aside from several case studies of methodological problems 
in field work among particular Asian/Pacific American subpopulations (e.g., 
Hurh and Kim, 1982; Yu, in press) , the systematic exploration of these kinds 
of methodological issues in surveys and other types of health research on 
Asian/Pacific Americans has not yet begun. 



(3) Problems in Interpretation of Correlates of Health Characteristics 

A substantive issue in using the NHIS data is the concern that 
differences in health characteristics among age strata may reflect either 
aging, cohort, or period effects, or possibly differences in the social 
composition of the respective age strata. Even a cursory analysis of the 
history of Asian immigration to the United States clarifies the scope and 
significance of this interpretive problem in analyses of Asian/Pacific 
American health data. 

As a result of differences in the waves of Asian/Pacific American 
immigrants, the interpretation of age variations is often confounded since age 
is associated with two other variables frequently used in data analysis: 
period and cohort (Schaie, 1965; Riley, 1973) . Awareness of the possible 
effects of age, period and cohort underlying variations in Asian American data 
is extremely important. Koreans of a specific age group, for instance, are 
likely to consist mostly of foreign-born persons while Japanese of the same 
age group will have a high proportion of native-born individuals. Because 
these two ethnic groups arrived in the United States at different periods of 
U.S. history, the age variable alone also implies differential occupational 
attainments between these groups. For Chinese and Filipinos, the ratio of 
foreign-born to native-born is slightly greater than 1.0, whereas for Koreans 
and Vietnamese the foreign- born far outnumber the native-born. Among 
Japanese, the proportion of foreign-born is relatively small compared to those 
of the other Asian American subgroups. 

Secondary analyses of NHIS data on Asian/Pacific Americans need to be 
cognizant of these complexities so as to avoid spurious interpretations of 
within group variations. For example, insofar as particular age categories in 
the NHIS subsample of Asian/Pacific Americans differ in their ethnic group 
composition, the analyst has to be wary of the kinds of historical and social 
crosscurrents which age-specific variations in health indicators might 
reflect. A careful assaying of national data, based on census and/or health 
statistics, in the context of a sociohistorical and sociodemographic 
understanding of specific Asian/Pacific American ethnic groups is much 
needed. Also needed is a specially designed study of the health of specific 
subgroups of Asian/Pacific Americans. 

In many ways, the methodological and analytical issues inherent in the 
NHIS data may also be found in another national data set maintained by the 
National Center for Health Statistics, i.e., the National Ambulatory Medical 
Care Survey. Findings from that data-collection system are presented below. 

261 



III. National Ambulatory Medical Care Survey (NAMCS) Data 

The NAMCS is a national probability sample of office-based physicians 
selected from master files of the American Medical Association and the 
American Osteopathic Association. Sampled physicians maintain a listing of 
all patient visits in their office during a randomly assigned 7~day period. 
The strength of these data is in the precision and depth of the medical 
information that it provides. Reliable data on diagnosis, reason for visit, 
diagnostic procedure, treatments, and medication therapy are reported by the 
physicians themselves. 



A. Limitations of the NAMCS Data 

Statistics from the National Ambulatory Medical Care Survey were derived 
by a multistage estimation procedure, which produces essentially unbiased 
national estimates and has three basic components: 1) inflation by 
reciprocals of the probabilities of selection, 2) adjustment for nonresponse, 
and 3) a ratio adjustment to fixed totals. Notwithstanding these safeguards, 
caution is warranted in the interpretation of the NAMCS data for Asian/Pacific 
Americans. First, ethnic identification is not made by the patients 
themselves but by the sampled physicians or their assistants, based on their 
prior observations or knowledge of the patients. Second, differences exist 
within the Asian/Pacific American population and between subgroups, as 
mentioned earlier. However, insofar as Asian/Pacific Americans tend to be 
viewed as "alike" and are treated similarly in the health care setting, it is 
legitimate for some data analysis purposes to examine their health care 
utilization behavior as a group. Third, due to the nature of the sampling 
design and the resultant disparity in sample sizes between specified 
race/ethnic groups, the estimates of the volume and characteristics of visits 
to office-based physicians are more precise for the majority white population 
than those obtained for Asian/Pacific Americans. Despite this, the NAMCS data 
for Asians and Pacific Islanders are sufficiently precise as to be useful in 
two important senses: 1) by way of contrast with similar data obtained from 
the white majority, and 2) by providing a preliminary estimate of the 
characteristics of the minorities sampled. 

B. Findings from the 1979 NAMCS Data for Asian/Pacific Americans 

The most recent analysis of the NAMCS data for Asian/Pacific Americans 
was made by Yu and Cypress (1982) . Their findings showed that in 1979, the 
majority of the physician visits made by Asian/Pacific Americans occurred in 
metropolitan areas (93.8 per cent). In contrast, less than three quarters of 
white American patients visits occurred in such areas. This pattern for 
Asians and Pacific Islanders is consistent with their geographic distribution 
in the United States. For instance, an independent estimate based on the 
1978 NHIS indicates that nearly 92 percent of Asian/Pacific Americans reside 
in standard metropolitan statistical areas. 

In the NAMCS data, no discernible sex differences were observed between 
Asian/Pacific Americans and the white majority with regard to visits. In both 
groups, proportionately more females (60 percent) than males (40 percent) 
visited physicians. In general, 45 per cent of the visits by Asian/Pacific 



262 



Islanders and 38 per cent of those by white Americans were for new problems. 
The ratio of return visits to visits for new problems was 1.2 for 
Asian/Pacific Americans and 1.6 for white Americans, a difference not 
statistically significant. Likewise, in terms of the kind of diagnostic and 
therapeutic services provided by the sampled physicians, no significant 
differences across specified race/ethnic groups were observed (Tables 4 and 
5) . Table 4 shows that the proportion of Asians and Pacific Islanders who 
received immunizations or desensitizations (6.3 percent) was not significantly 
different from that for white Americans (5.4 percent). Despite the stereotype 
that Asian women have a strong sense of modesty, the same proportion of Asian 
and Pacific Islander as white American women (5 percent) received Pap tests 
(Table 5) . Insofar as blood pressure checks are concerned, slightly fewer 
Asian/Pacific Americans (33 percent) received this diagnostic service than did 
white Americans (35 per cent) . 

There are, however, significant differences between the groups as far as 
the age distributions of the patients are concerned (Table 6) . For 
Asian/Pacific Americans, as many as 30.4 percent (roughly 1.7 million divided 
by 5.6 million) of the estimated visits to physicians' offices were made by 
children under 15 years of age, which is significantly different from that 
observed for whites (18.2 percent). The visit rates calculated for the two 
groups are also shown in Table 6. First, compared with white Americans, 
Asian/Pacific Americans of all ages have a lower rate. Second, while the 
visit rate for whites shows an almost linear increase with age, that for 
Asian/Pacific Americans displays a more U-shaped relationship with age. 
Third, within-group comparisons indicate that among Asian/Pacific Americans, 
in contrast to whites, the youngest age group had the highest visit rate. 

In terms of medical specialty, about one-third of the visits for both 
white and Asian/Pacific Americans were made to the offices of general and 
family practitioners (Table 7) . However, significantly more visits by Asians 
and Pacific Islanders (19 percent) were made to the offices of pediatricians 
than were those of whites (10 percent) . These data are, of course, consistent 
with the earlier finding that a sizable proportion of the physician visits by 
Asian/Pacific Islanders were made by patients under 15 years of age. 
Dermatology is another specialty where a statistically significant difference 
exists between Asians and white Americans. In addition, a significantly 
smaller percentage of Asian/Pacific Islanders made visits to the offices of 
general or specialty surgeons or psychiatrists. The cultural underpinnings of 
these findings will be discussed later. 

Table 8 presents data on physician visits by major ICD-9 Diagnostic 
Codes. Only 1.9 percent of the visits made by Asian/Pacific Americans were 
for mental disorders compared with 4.5 percent for whites. The difference is 
statistically significant. In contrast, significantly more visits were made 
by Asian/Pacific Islanders (10.2 percent) than by white Americans (5.3 
percent) for problems related to diseases of the skin and subcutaneous 
tissue. Examination of the more detailed ICD-9 codes for this category 
reveals that contact dermatitis and other eczemas, as well as diseases of the 
sebaceous glands, are two diagnoses in which the visits are proportionally 
higher among Asian/Pacific Americans than whites. Since these diagnoses 
include diaper rash and acne, their relative preponderance is most likely an 
artifact of the overrepresentation of visits of children under 15 years old 
among Asians and Pacific Islanders. 

263 



Table 8 also shows that 23 per cent of the Asian and Pacific American 
visits, compared with only 16 percent for the whites, were made for the 
category "special conditions and examinations without sickness." This 
diagnostic group includes general medical examinations, routine normal 
pregnancy (prenatal) examinations, and supervision of healthy children 
(well-baby examinations) , which together constitute the largest proportion of 
visits for specific diagnoses regardless of race. Of further interest is the 
significantly smaller proportion of visits for injury and poisoning observed 
for Asian and Pacific Islanders (5.3 percent) than for whites (9.3 percent). 



C. Cultural Diversity Reflected in NAMCS Data 

In many ways, the results of our analyses are both heartening and 
disheartening. On the one hand, the visit rate for Asian/Pacific Americans is 
rather low in every age group except for those under 15 years compared with 
that found for white Americans. A persusal of NHIS unpublished data lends an 
interesting interpretation to the NAMCS finding. From 1976 through 1978, 
visit rates to emergency room/outpatient clinics were significantly higher for 
Asian/Pacific Americans (798.1 per 1,000 population) than for white Americans 
in 1978 (566.0 per 1,000). This NHIS finding suggests that the NAMCS data 
give only a partial picture of the use of health services by Asian/Pacific 
Americans. It is most likely that language and cultural barriers have 
prevented many immigrant Asians and Pacific Islanders from having a regular 
family doctor, as reported earlier in a few localized studies (Lieu et al. , 
1972; Weaver, 1976). Lack of health insurance, ignorance about the service 
delivery system in the United States, and a sizable number of unattached 
adults and elderly in the Asian populations are three other possible factors 
in accounting for their heavy reliance on emergency room and/or outpatient 
clinics. 

On the other hand, at least one reason for optimism exists. Taken at 
face value, the NAMCS data fail to show gross disparities, across ethnic group 
membership, in the kinds of diagnostic and therapeutic services provided by 
the sampled physicians. Instead, some interesting cultural diversity in 
medical care is reflected in the NAMCS data. For instance, within the 
Asian/Pacific Islander population, compared with the other age groups, 
children under 15 years of age had the highest rate of visit to office-based 
physicians. We suspect that this may be due to the presence of a sizable 
proportion of persons in the under 15 age group, many of whom are children of 
immigrants. 

As a result of the recency of their arrivals, we can expect that the 
majority of Asian Americans will continue to retain many of their traditional 
values and health practices in the old countries, where health is viewed as a 
state of equilibrium ("homeostasis") between man, society, and the cosmic 
forces of the universe. Maintaining a balance between the "hot" and "cold" 
elements of the body is the cornerstone of good health. Disease is a 
disturbance of that relationship. Blood is a source of human vitality and is 
difficult to replenish. Consequently, surgery is avoided to the extent 
possible since it increases the risk of losing one's blood. Indeed, in the 
1978 National Health Intevview Survey, 93 percent of 919 Asian/Pacific 
Americans in the 17-64 age range had not given or sold blood in the past 



264 



year. Seventy-seven percent of those interviewed had never done so. These 
concepts, of course, are so deeply rooted in the traditional cultures of 
several Asian societies and they may cross ethnic boundaries. Some of the 
similarities in herbal prescriptions (e.g., the use of Ginseng for restoring 
vitality or stress endurance) and behavioral proscriptions (e.g., never 
sleeping with wet hair; "doing the month" after delivering a baby) are rather 
striking (Popov and Goldwag, 1973; McKenzie and Chrisman, 1977; Gould-Martin, 
1978; Pillsbury, 1978; Sich 1981) . Taking into consideration the composition 
of the major Asian/Pacific American groups, i.e., Chinese, Japanese, Korean, 
IndoChinese, and Filipino, it is apparent that historically at least three of 
these groups have been heavily influenced by China — in language, religion, 
philosopy, social structure, and medicine. The fact that the Chinese presence 
in the Philippines began as early as the 11th century suggests the possibility 
of its influence in Filipino society — though not to the same extent as that 
found in the other countries. Therefore, despite the apparent cultural 
diversity among Asians, unifying threads of social values and health beliefs 
are interwoven into the indigenous cultures. The persistence of these 
health-related beliefs and practices in Asian American subcultures is evident 
from the few studies in the United States based on limited samples and 
sometimes impressionistic observations (Campbell and Chang, 1973; Ling, King, 
and Leung, 1975; Chan and Chang, 1976) . Obviously, precise empirical evidence 
about the prevalence and sociodemographic variations of these ethnic practices 
are lacking. 

In the NAMCS data, demographic factors probably played an important role 
in the high physician-visit rate of Asian/Pacific Americans for "special 
conditions and examinations without sickness," while cultural factors may be 
salient in their unusually low proportion of visits for injury and poisoning. 
The former is essentially preventive health care, while the latter, though 
usually unanticipated, is preventable in most cases. Thus, the juxtaposition 
of these two findings would tend to suggest that although proportionally fewer 
Asian/Pacific Americans rely upon office-based physicians for their regular 
source of medical care, many of those who do visit physicians are apparently 
prevention-oriented. It is further possible that this consciousness for the 
importance of prevention may extend beyond the supervision of infants. 



IV. Mental Health Status Indicators 

In general two types of statistics, treated and true prevalence rates, 
are used conventionally to provide information on the mental health status of 
a given population. The 1980 Census provides some information on treated 
prevalence for selected ethnic groups based on estimates from a sample, 
whereas true prevalence data are obtained from psychiatric epidemiologic 
surveys conducted in the community by means of either a standardized 
diagnostic instrument or at least a symptom-rating scale. In what follows, we 
will review the census data on inmates of institutions in the United States in 
order to provide estimates of the treated prevalence rates. Additionally, 
findings from recent small-scale studies will be reported to provide as much 
information as possible on the true prevalence rates for Asian/Pacific 
Americans. 



265 



A. Treated Prevalence Rates 

Among the institutions relevant to mental health issues, only mental 
hospitals, correctional institutions, and homes for the aged have been 
tabulated in published census reports. In a recent analyis, Liu and Yu (in 
press) found that in every case, for males as well as for females, the rates 
for Asian/Pacific Americans are lower than that found for white Americans. 
For inmates of mental hospitals, the age-adjusted rate for white American 
males is 1.20 per thousand, for females, 0.70; for Asian/Pacific American 
males, 0.45 per thousand, for females 0.24. The white-to-Asian ratio among 
males is 2.67 and among females, 2.92, indicating that at least two-and-a-half 
times more whites than Asian/Pacific Americans are committed to mental 
hospitals in this country. 

For correctional institutions, the age-adjusted rate for white males 
(2.33 per 1,000) is twice that for Asian/Pacific Americans (1.11 per 1,000) . 
Among females, the difference is smaller, the rate for whites being 0.14 and 
that for Asian/Pacific Americans 0.08, yielding a ratio of 1.75. 

In homes for the aged, the Asian rates are low as well. For males, the 
white rate is 2.72 per thousand and for Asian/Pacific Americans, 1.49 per 
thousand, the ratio being 1.83. For females, the white rate is 3.72 per 
thousand and the Asian/Pacific American rate 1.38 per thousand, yielding a 
ratio of 2.70. 

The low institutionalization rates for Asian/Pacific Americans should be 
interpreted with caution, because more than half of them are immigrants. 
Selective factors which determine the types of persons who are allowed to 
immigrate to the United States mitigate against the likelihood of their 
commitment to institutions. Furthermore, some evidence is available that 
Asian/Pacific Americans tend to return to their parent country for treatment 
of mental illness rather than face confinement in the U.S. where service 
providers are less familiar with their cultural conflicts and life stresses 
(Yeh et al., 1979) . 



B. True Prevalence Rates 

Data based on institutionalized populations do not provide useful 
information on the true prevalence of mental disorders in the community 
because selection factors determine who among those suffering from a 
particular type of psychiatric disorder actually receive treatment (Dohrenwend 
and Dohrenwend, 1969: 5-7). The problem is compounded by the lack of a 
standardized method of case-finding that can be used in a uniform and 
consistent fashion in population surveys to detect persons with mental 
disorders (Kramer, 1976: 188) . But the development of case-finding techniques 
depends on the existence of a consensus, which was lacking, among members of 
the psychiatric profession as to what constitutes "psychopathology, " "mental 
illness," or "psychiatric disorder." The publication in 1980 of the 
Diagnostic and Statistical Manual , Version III (or DSM-III for short) by the 
American Psychiatric Association represents one of several attempts which 
started in the 1960s to develop a consistent definition of mental disorders. 
The release in 1981 of the Diagnostic Interview Schedule (called DIS) , Version 



266 



Ill (Robins et al. , 1982) represents a new case-finding technique based on the 
DSM-III criteria that can be used in large-scale community surveys. This 
technique has been used to collect information in such surveys on white, 
black, and Mexican Americans in certain parts of this country (Regier et al., 
1984) . Overseas, the DIS has been used in several independently organized 
surveys in Taiwan, Mainland China, Hong Kong, Korea, and Peru. The results of 
these studies, when analyzed and published, should provide the best possible 
estimates to date on the true prevalence of mental disorders in several 
countries, as well as establish the reliability and validity of the 
DIS-Version III in community surveys both in the United States and across 
cultures. Until then, we are forced to use data collected in smaller studies 
to ascertain the mental health status of Asian/Pacific Americans. 

Psychiatric epidemiologic studies designed to provide data on the true 
prevalence of mental disorders in community settings fall into two broad 
categories: (1) those that yield information by means of clinical diagnoses 
based on the use of a diagnostic instrument or clinical judgments rendered by 
professionals (usually psychiatrists) on symptom data elicited through survey 
interviews, and (2) those that yield psychiatric symptomatology or 
psychological distress data but not clinical diagnoses; these studies employ 
actuarially calibrated symptom scales to screen persons who are "cases," i.e., 
those who score above a cut-off point which was either pre-specif ied by 
previous validity studies with clinic patients or determined from newly 
studied samples of human subjects. Both sources of psychiatric epidemiologic 
data continue to suffer from serious methodological problems, but they provide 
the best information yet available for understanding the rates of mental 
disorders outside the institutionalized context. 



(1) . Community Surveys using Clinical Judgments 

Unfortunately, diagnostic data on Asian/Pacific American populations 
using the most recently developed instrument in the United States, the 
Diagnostic Interview Survey-Version III (DIS-III) , have not yet been collected 
on a community-wide basis. However, a pilot study to investigate the 
feasibility of using such an instrument on Chinese Americans has been 
conducted recently as part of a collaborative project between the 
Pacific/Asian American Mental Health Research Center and a primary health care 
clinic in New York City. Our experience indicates that the DIS, with some 
modifications, can be applied to Chinese Americans in an interview situation. 
In this primary health care setting, the response rate was better than 95 
percent. Men as well as women showed little hesitation in talking about 
mental health problems when interviewed in a health help-seeking context. 

Table 9 summarizes the preliminary findings from that study. Of the 342 
patients interviewed at the clinic, the proportions that manifested symptoms 
that yield the clinical diagnoses of Anxiety Disorder using the DIS/DSM-III 
Criteria were: 7.7 percent of those 18-24 years old, 15.9 percent of those in 
the 25-44 age group, and 20.4 percent of in the 45-64 age group. Among young 
adults, more men (11.4 percent) than women (4.6 percent) exhibited clinical 
anxiety. After age 25, there is a sex cross-over such that more women than 
men expressed clinically severe anxiety symptoms. 



267 



About 14 percent of those in the 25-44 age group reported symptoms that 
are clinically recognizable as Somatization Disorder. The next highest 
proportion {9.1 percent) is to be found in the 45-64 age group, followed by 
3.8 percent among the young adults. For each age group, the rates for women 
are higher than those for men. 

Panic disorders, like anxiety disorders, are reported by more persons 
(9.1 percent) in the oldest age group, 45-64 years, than in the other groups. 
More women (12.9 percent) than men (7.1 percent) in that age group had the 
disorder. 

In younger ages, more men than women reported clinically severe panic 
symptoms. But the number of cases is far too small for us to have confidence 
in the meaning of these figures. 

The percent of persons having had a Major Depressive Episode also 

increases with age. In the two younger age groups, more women have reported 

having such a mental health problem, but in the 45-64 age group, sex 
differences were reversed. 



(2) Screening Scale Studies Using Cut-Off Scores to Identify Cases 

Although there have been numerous studies using screening scales to 
identify likely "cases" of mentally disordered persons, not all have collected 
information on race or ethnicity (e.g., Manis et al. , 1964; Phillips, 1966). 
Of those that have, findings on non-white populations in the sample are not 
always analyzed separately. In the 30-year period between 1950 and 1980, only 
15 screening scale studies can be found in the literature that report findings 
on race or ethnic differences in "caseness." None contains data on 
Asian/Pacific Americans. 

At present, however, a widely-used screening scale for depression, the 
Center for Epidemiologic Studies-Depression Scale (CES-D) , is being pilot 
tested on Chinese Americans in our New York clinic study. The findings are 
shown in Table 9. Using a cut-off of 16 and over to determine caseness, the 
data indicate a gradual increase of symptomatology with age. In the young 
adult group, there are hardly any sex differences in depressive 
symptomatology. But with increasing age, women report more depression 
symptoms than men. These symptoms may not be clinically severe, however. 

Chinese American responses to the Demoralization Scale recently developed 
by a University of Columbia team of researchers are similar. Sex differences 
are practically non-existent in the young adult group, but are found in the 
older groups. Further analyses of these data are being conducted to determine 
some of the socio-demographic factors associated with the various types of 
clinical disorders and depressive symptomatology. 

One study has reported on the use of the CES-D in a community sample of 
Asian Americans (Kuo, 1984) . The mean CES-D score for Kuo's Asian American 
sample is higher than the means of the white samples reported in other studies 
(e.g., Radloff, 1977; Frerichs et al. , 1981). A greater proportion of the 



268 



Asian sample (19.1 percent) had a depression score of 16 or above, which is 
higher than the rates previously found for whites. Among Asian Americans, 
statistically significant differences were found for Koreans, Filipinos, 
Japanese, and Chinese, even after holding constant several demographic 
variables (sex, marital status, age, nativity). Kuo's study remains to be 
replicated in other locales, using a more refined method for identifying and 
sampling Asian ethnics. 



C. Mental Health Services Utilization 

To date, the only published data on services utilization for 
Asian/Pacific Americans on a national level are those reported by Yu and 
Cypress (1982). Table 7, shown earlier, reveals the significantly small 
percentage of visits made by Asian/Pacific Americans to the office of a 
psychiatrist. Moreover, regardless of the medical specialty of the physican 
they visited, only a very small percentage of Asian/Pacific American patients 
received the principal diagnosis of "mental disorders." Thus, the NAMCS data 
lend confidence to previous observations made by practicing psychiatrists and 
clinical psychologists. To illustrate, in a study of 17 community mental 
health services in the greater Seattle area. Sue (1977) found that over a 
period of three years, only 3.1 percent of the Asian American clients saw a 
psychiatrist at intake, and not one Asian saw a psychiatrist during therapy. 
A number of researchers believe that cultural factors play an important role 
in the Asian reluctance to consult with psychiatrists, even when they admit to 
having psychological problems. For the most part, Asians and Pacific 
Islanders appear to have a tendency to somaticize their psychiatric symptoms 
(Tseng et al., 1974). Hence, the consensus among practitioners is that 
members of this minority group generally perceive "talk" therapy to be 
ineffective (Tung, 1980; Chien and Yamamoto, 1981) . Instead, self-medication, 
such as herbal tonics, is the expected treatment even for mental problems, 
thereby causing an extensive delay in help-seeking until the disorder has 
clearly become unbearable or unmanageable. Lin and Lin (1978) reported from 
their experience that the delay among Chinese patients in seeking psychaitric 
help can be as long as twenty-five years. Intrafamilial resources are first 
utilized before outside help is sought and, even then, respected 
intermediaries who have the family's trust are consulted before assistance is 
sought from a psychiatric professional, who is a total stranger to the 
family. These findings, consistent as they are, are not based on large-scale 
epidemiologic studies. Hence, until prevalence and incidence data on both 
"treated" and "untreated" mental disorders among Asian/Pacific Americans are 
available, the extremely low percentage of visits to psychiatrists by this 
growing minority must be interpreted with caution. 



V. Summary 

In summary, we have presented the major findings on the physical and 
mental health status indicators of Asian/Pacific Americans taking into 
consideration the limitations of the data. 

Physical health status indicators in the NHIS data indicate that, 
compared to the total U.S. population, a smaller proportion of Asian/Pacific 



269 



Americans reported being in "fair" or "poor" health, or having physical health 
limitations. Even so, approximately one out of five reported having had no 
regular source of care, and the visit rates of Asian/Pacific Americans to 
emergency room/outpatient clinics were significantly higher than that reported 
for white Americans. Data on office visits collected by NAMCS suggest that a 
significantly small percentage of Asians or Pacific Islanders had visited a 
physician for injury and poisoning; a substantial proportion of their visits 
were apparently made for preventive care: general medical examinations, 
routine normal pregnancy examinations, and supervision of healthy children. 
Significantly fewer visits were made by Asian and Pacific Americans to the 
office of a surgeon or a psychiatrist. A cultural resistance to the 
utilization of these two types of medical specialty is suggested. 

Some traditional concepts of health rooted in the Asian culture appear to 
persist. This is evidenced by the extremely low percentage of Asian/Pacific 
Americans who have ever donated blood. Other interesting cultural influences 
are discernible in the health practices of Asian/Pacific Islanders. Seven out 
of ten Asian/Pacific American adults had never smoked or had quit smoking. 
More than four-fifths of this population either never drank alcoholic 
beverages or do so only occasionally. 

Although much of the data reported here are based on small samples, we 
are impressed by the consistency of findings from independent sources of 
data-collection systems. By and large, these findings support intuitive 
impressions that such risk factors as smoking and drinking are indeed less 
prevalent among Asians than among the majority Americans. 

A review of mental health status indicators reveals that, compared to 
white Americans, Asian/Pacific Americans have lower rates of commitment to 
mental hospitals, correctional institutions, and homes for the aged. However, 
a pilot study based on patients entering a primary health care setting show 
one Asian subgroup, Chinese Americans, to have high rates of major depressive 
disorder, generalized anxiety, depression symptoms, and demoralization. In 
addition, a community study using the depression symptom rating scale suggests 
that Asians have higher mean depression scores than reported for whites in 
previous studies. 

While the validity of these diagnostic instruments and screening scales 
has yet to be firmly established for Asian/Pacific Americans, and their 
reliabilities remain to be tested, there is sufficient evidence to indicate 
promise in the development of standardized instruments applicable to studies 
of white and Asian Americans. The extent to which the disclosure of mental 
disorders or depressive symptomatology among Asian/Pacific Americans is 
influenced by cultural attitudes toward mental illness has not been studied. 
Such attitudes may well turn out to influence definitions of behavior 
disorders with consequent disparities in statistics on mental illness. 

Certainly, more systematic large-scale studies are warranted on the 
comparative health and mental health behavior of Asian/Pacific Americans and 
other ethnic groups, especially white Americans, in order to increase our 
understanding of intra- and inter-ethnic differences in health and mental 
health. 



270 






4) 




4-1 

S.S 


tu 


E 




CO 


u 


1 — o 
*j CL in 





U in -rt _ >v 



1-^ -p-l -, 





•M 


>^ 


c 


w 


■l-t 


o 


'> 



C 1- 

.-I 4-1 g (U 

— H in ^ 

in -^ — 4-> 

>, > R in 






Ci. 



0) u ^^ 
> O to 

.-< lU 

(U i-i x: 
o —I 
u n u 

(U UJ o 

cu o 
a, 



to Uj > 

4J O -rt 

e o 

— I cd 



X 
0) 

in 
•a 



c/5 



C/3 



J3 
CO 



J3 



OO 1— 1 O vO 



^T vO O CT» 
O O ^ »M 



oo LO r- o 
to oo o r-j 



Lrt to to 1-H 

t-H »-H r^j ^ 



■«T LO -"a- '^ 

o> OO »-< c^ 
rr rr in rv| 



rj oo tn L/i 
to r^ •— I rg 
r- Oi rO i-i 



Oi vO ir» to 

O O 1-H to 



GO to rO LO 

to ^- r->" r^ 



O to CM to 
rj rj to -^ 



r^ o to ^^ 



Lrt «0 to m rH 

i-H ,-H to LTl .-H 

to r^ .-H to r-i 



OO O O ^O 



0^ ^T •— I '^ 
to rJ sD in 



00 in »-H »— 1 TT 

o 1—1 1—1 rg in 



»o oo r~- 1~ 


a> 


O ^0 -^ 1-H 


^ o .-4 r-S 


d 


r4 --5 rg ro 






>o 


0» CO O ^9" 


d 


d d .-H ro 


to 


1-* l/^ LO to 


en 


^ oo r-^ ro 


(O 


lO vO VO M 


d 


d d 1-5 lo 



00 OO OO lO 



in ,— ( \0 to 



lo i-< in ^ 

1— 4 T-l 1-* to 



1— ( r-i 1— I o 
^T r^ OO in 



\0 OO 1-t CNJ 

o o (Ni ^r 



to vo r^ in 
csi in in oi 



rg r^ in o 



so to to r^ 
1-4 1-H r4 -^ 



o ^S" CO ^a- 



r*- OO o p^ CTi 

O O t-4 r-H ^ 



r-^ o p^ 00 to 



^r fsi in a> to 

O O O 1-4 rr 



^ CM o m 



<s 



C>0 
CO 



c r^ in in 

5 «-< -^ o 



<i> ^ o >x 
c r^ in in 

3 t— I ^iT o 



0) 

OO 



•o I I 

e r*- in I 
5 •-< -^ ^ 



T3 
<L> 



J3 
D. 



> 



a 
<a 

X 



c 



u 

c55 



o 



c 



o 



ra 
o 

c 



f2 



Xi 



ia5 
xco 

•a > 

c « 
.-. u 

CO (1) 

^-' Q. 
jll in 
O U 



0) 

O0 4-> 

to o 

4-1 3 

m o 

U I- 

u a, 
1) 
a, o 

3 OO 

c 

V- -^ 

O TS 
U-i XI 

to 
to c 

> OJ 



3 



V 

t_> o ^-» 
C Oi 

V . [U 

T3 ,-1 

.-I C/J 

tiJ >. •— 

C XI 

o 
u <o 

DO 
4-> to 
C 4-> 

a) c 

U 0) 

t> u 

V u 
D. (U to • 
1 p. T3 -a 

in CO 

Oi 4-t rt J3 

to 4-' 4-1 

to -M x> E 



V) 



u 



271 



Table 2 



Selected health practices of Asian/Pacific Americans 
20 years and over: unweighted data 



Practice 



Number 



Percent 



Asian/Pacific Americans 20 years and over- 

Frequency of breakfasting 

Every day 

Sometimes 

Rarely or never 

Frequency of snacking 

Every day 

Scmetimes 

Rarely or never 

Number of hours usually sleeps 

6 hours or less 

7-8 hours 

9 hours or more 

Perceived physical activeness 

More active than others 

As active 

Less active 

Smoking status 

Never smoked 

Present smoker 

Former smoker 

Frequency of most frequently used 

alcoholic beverage 

Never use alcohol 

Occasionally 

Once or twice a week 

Three or more times a week 

Obesity status 

Obese^ 

Non-obese 



256 



144 
54 
56 



85 
96 
75 



52 

178 

25 



88 

136 

29 



154 
73 
27 



81 

132 
22 
21 



32 
220 



100.0 



56.7 
21.3 
22.0 



33.2 
37.5 
29.3 



20.4 

69.8 

9.8 



34.8 
53.8 
11.5 



60.6 
28.7 
10.6 



31.6 

51.6 

8.6 

8.2 



12.7 
87.3 



Source: National Center for Health Statistics, 1977 National Health Interview 
Survey, 1/3 subsanqsle of persons 20 years and over, unpublished data. 

^Obesity is defined here as a body mass index (weight/height ) of 27 or greater 



for males; 25 or greater, for females, 
and height in meters. 



Index was computed with weight in kilograms 



272 



Table 3. Characteristics of regxilar sources of health care 
among Asian/Pacific Americans: unweighted data 



Characteristic 



Number 



Percent 



All Asian/Pacific Americans 

Regular source of care 

Yes— - - - — ■ 

No- - 

Unknown 

Usual place of care^ 

All with a regijlar source 

Doctor's office 

Hospital clinic or emergency room 

Health center 

Other 

One particular doctor^ 

All with a regular source 

Yes 

No- — 

Unknown 

Time to get to regular source 

All with a regular source 

Less than 10 minutes 

10-14 minutes 

15-19 minutes 

20-29 minutes 

30-44 minutes 

45-59 minutes 

60 minutes or more 

Unknown 

Reason for not having a regular source 

All without a regular source 

Haven't needed a doctor 

See different doctors for different problems- 
Former doctor not available 

Haven't found the right doctor 

Just moved here 

Other reason 

Unknown 



1,456 



1,175 

269 

12 



1,175 

968 

124 

29 

54 



1,175 

876 

252 

47 



1,175 

219 

224 

220 

191 

166 

49 

62 

44 



269 
147 
33 
15 
16 
32 
23 
3 



100.0 



80.7 

18.5 

0.8 



100.0 

82.4 

10.6 

2.5 

4.6 



100.0 

74.6 

21.4 

4.0 



100.0 
18.6 
19.1 
18.7 
16.3 



14.1 
4.2 
5.3 
3.7 



100.0 

54.6 

12.3 

5.6 

5.9 

11.9 

8.6 

1.1 



Source: National Center for Health Statistics, 1978 National Health Interview 
Survey, unpublished data. 

^is information was obtained in the interview only for persons with a regular 
source of care. 



273 



u 

O 



Ul 

0) 
•D 
C 
(U 

u 

tn 
o 

> 

u 
(U 

ca 



lU 

0) 

sz 


(0 
XJ 





^3 


fll 


XI 


X 


-w 


c~^ 


C 


jj 


3 


>i 




£i 


• • 

OJ 


(0 


U 


> 


0) 


c 
u 


-r-( 

u 

2. 


« 


w 


>, 




^ 




Cli 




XJ 




c 




<u 




o 




u 




01 




&C 




'^ 




GJ 




iH 




JD 




Q 




Eh 





en 
%. o c 



0) 
XJ 



en 



r-l 



OJ 

o <u 

XJ C 
3 0) 

03 cn 

u 0) 
0) O 

JZ --H 

Eh > 
u 
0) 

in 



^ o 
n • 

in o 
•■o 

m 
ir> 



O^ O CM 

fO • • 

CO o o 

-O CM 

VO ■— I 
CM 

CM 

O 

in 



iH O 

m . 
^T o 

•■ o 

n rH 

in 
in 



CO 



XJ J 

ra c 
t-' XJ u 

tn Ul cu 
XJ H fi. 



in 



in 



o 
in 



in 



en 






(TV 
W3 



0^ 



CM 






in 



CO 

in 






in 



o 

vo 



CM 
CM 



«3" 

9 



in 

CM 






CO 
CM 



n 


O 


vo 


vo 


VO 


• 


• 


• 


• 


• 


CM 


n 


,r^ 


«3" 


n 


CM 











in 

CO 



1 
c 












1 

i 
1 








o 












I 








c 












1 
1 


u 






T! 








1 




1 


XJ 






C 












1 


Z! 






(a 












1 
1 


^ 






c 








1 




I 


ra 






o 








1 c 


yt 


1 


i_i 






•»-t 








Z 1 £ 






OJ 






XJ 


1 






D 1 


H 


1 


JZ 






CL 








-1 cn en 1- 


H 




E- 






•<-( 


"c 






J C C ( 


J 


>i 


\ 


1 




u 


o 




\,' 


a -^ ■•-i t 


n : 


>i >-> 


>, 


1 : 


>i 


O 






C t 


M .-i C ( 




i 0) 


C 


1 c 


1, 


w 


X) 




-2 " 


hoc: 


5 1 


ai 


ro 


(n f 


a 


<u 


04 






-I to (a ( 


3 J 


J U 


iu 


C I 


-( 


t-l 


— 1 t 


~ 


XJ •. 


^^ C r-^ I 


V 3 


QJ 


■-H < 


ij 


a 


S-l ( 


D 


ro t 


n ~ p. 


J 


: tn 


x: 


C J 






u 


H 


N 


= O 


-H a 


J 


XJ 


0) 4 


J 




tn J 


J 


• rH 


D U > 


13 ( 


3 (U 


O 


XJ 




4) en 


OJ ( 


J 


c 


n r— : ( 


J 


H U 


s: 


tn 1 


-J 


U ( 


D 




D XJ — 1 -. 


H t 


n --H 


u 


■^ ( 


i; 


c 3 


a, -r 


-1 


g ^ 


3 0) E 1 


3 : 


>i li-i 


> J 




O u 


( 


~ 


^ 


••H (0 < 


U J 


- U_l 


tn 


-1 


J 


2 C 


1- 


-\ 


l-l 


Q t : 


•: c 


u O 


e. 


C 


3 



01 
(U 
XJ 

ro 

XJ 

w 
•a 

0) 



c 

QJ 

x: 

XJ 



w 
c 
ro 
J.; 
tn 



c 
ta 

tn 

c 

(0 

■n 

c 



u 
O 
«-i 

ra 

XJ 

ra 
•a 

in 
o 

I—! 

o 

c 



Ul 

Si- 
's 

Wl 

<u 

c 

0) 

t-l 



0) 

> 

u 

0} 
CQ 

OJ 

c 
o 

c 
ra 



QJ 
Ul 



QJ 

tn 

D 

ra 
o 



^ ^ 



ra 

XJ 

o 

XJ 

O 

c. 






o 

XJ 

c 
o 
u 

U: 

CI 
Cl, 
CM 



c 
ra 



w 
>i 

x: 
Oi 



274 



o 



0) 

0) 
13 
C 
QJ 
1-1 

V) 
CD 
U 
-H 
> 
U 

w 
o r~ 

■■H CT* 

J-> i-H 

in 

o " 
c w 
en 0) 
nj oJ 

-H IB 

W 
u-i 
O 'O 

0) 
m 4J 

Qj-i-l 

S C 
JJ O 

^.. 
.0) 

tn u 
tn 



■s 



c «-l 
(D -'< 

-H Ul 

CO (0 

jJ 

C 
0) 

u 

1.1 

0) 



in 
(1) 

(TJ 



^ 



03 
U C 

■-r-i n3 
U-I u 

ra 0) 
5 



3 






W 
Q) 
O 

> 

0) 

U U 
-.-I 0) 
JJ T3 
03 C 
O 0) 
C M 

cn 

Q 



■o- 


o 


CM 


CO 


ro 


O 


as 


CN 


rs) 


in 


U5 


rsi 


• 


• 


• 


• 


• 


• 


• 


• 


• 




ID 


o 


CO 


'a* 


O 


in 


.— 1 


IT) 


rn 


o 


VD 


w 


o 




\o 


tN 




CM 




ro 






CTi 


—1 




















in 






















LT) 























CO 



o 



ro 



in 



<y\ o 
ro « 
CO O 

CM 
O 

in 



o 



ro 






o 
in 



ro 
CN 



CO 



ro 



in 
ro 



CM 



rsj 



ro 



>-l o 
ro • 
^r o 

ro rH 
r-l 

ro 

VD 

in 
in 



ro 

V£> 



CO 
vo 






CN 



ro 
(N 



CN 

CO 



VD 

ro 



rsi 



o 
vo 



in 



in 
ro 



w 




>1 


! 

1 






J 


£ 


4J 




u 


u 






8 


-H 




O 


O 






03 




JJ 


JJ 


J 


J 


jr 


^ 




03 


03 


0) 


u 


> 




— ^ 


•H 


0) 








fa 


.C 


JJ 


0) 


U-I 

O 




^ 


ro 


07 


org 


X 


X 


fH 


03 


J-> 4J 


0) 


01 




<U 


r-l U 


-s 


-i^ 5 


s '^ 


1-1 
Oi 


JJ M 


JJ 


1-1 J 


J -w : 


r 1 


03 Q) i 


J —I 


0) 


c » 


m &4 1 


3 J 


C ! 


i -H I 


c 


S I 


n-t 


r-l 




s 


3 J 


U c 


J > 


< Q 


Q 



01 
0) 

g 



Q 



I 

0) 

-8 



03 

cn 
ro 



1-1 



03 
03 
JJ 

ro 
JJ 

T> 
0) 
jJ 

c 

03 

D 
O 

c 

E 
ij 

0) 



a; 

x: 
JJ 



03 

03 

ro 

13 



03 
C 
lU 

•F-l 

13 

C 



1-1 

o 

uj 

ro 
jj 
ro 

13 

w 

(1) 

'B 

o 

C 



'S 

1-1 
0) 
13 
C 
0) 
u 



0) 

u 

-rH 
> 

1-1 

0) 
03 

Q) 

C 

o 



8 § 



JJ 



1-1 



0) 

03 

3 

ro 
u 



ro 
JJ 

Q 



13 
13 

ro 



c -^ 

oJ a. 

^ u 



03 



275 



CTl 



w 

CD 
J-) 

ra 

X) 
U3 

ID 

a> 

•H 

c 



0) 

u 



O 
M 

O 

CD 

Cn 

m 



o 

M 
>i 

x: 

5 

03 

-H 

W 



U-l 

O 

01 
0) 

XJ 

nj 
u 



D 



o 

ro 
a, 



^ 



0) 



03 

u 

(0 



4-> Hc 

•i-l 0) 

in 4J 

> « 



U3 

0) 

4-) J-) 



■r-l 
> 



03 



4J * 

U3 -U 

-H to 

> « 



03 


03 



-u * 

-W Q) 
03 4J 

•rH (0 

> e2 



4J 

•1-t 
03 
•l-l • 

> 



03 



03 

w 



in 


03 


<y\ 


(T> 


-H 


T 


in 


cn 


cn 


•^ 


in 


CTl 



CN 

in 

CTl 

in 
in 

in 



o 



OJ 



n 

03 
CM 

cri 

CN 
O 

in 



CN 



ro 

n 
ro 

in 
in 



o 
o 
CO 

o 
•J^ 

J3 



ro 
CN 

in 

VD 



o 

ro 

in 
o 
in 



ro 
ro 



ID 

o 



<X5 



CN 
CN 



o 

03 

r«- 

CN 



cy> 



CO 



CN 



CM 



O 



ro 



in in Tf 

CO CO VD 

CN •<S' rH 

•• » » 

in r^ o^ 

cTv CO in 

CN ro 00 

«> w ^ 



ro 



00 


rv) 


in 


a\ 


VD 


VD 


in 


o 


o 


in 


(Tt 


CM 



CM 



CN 



CN 



CN 



CO CN CM cr, 

CTJ CO iH crt 

<N r« CT» CN 

•• V k » 

CN <TV 00 •^ 

m CO iH CTi 

ro CN r~ in 



CN 
CO 



m 



CO 

CN 



O ai 



O 



-O 



03 
(U 



CD 
0) 

I 



Q4 

O 

M 



0) 

cn 



03 

>1 
in 



C 



03 
CN 

in 



03 

>1 

I 

in 

CN 



03 
VD 



276 



U 

to 

'3 



u 




K 




en 




u 




O 

U-l 




>i 




JJ 




iH 




(0 








s. 




en 


CT\ 




r~ 


iH 


CT\ 


m 


<-i 


o 




-1-1 


^ 


'O 


m 


GJ 


0) 


£ 


4J 




m 


>i-i-i 


ja 


CO 


tn 


T3 


4J 


Q) 


-■-( 


JJ 


w 


-^ 


-rH 


c 


> 


3 



o 

w 

r; 

04 



c 

o 
u 



o 

(0 



m 



u-i u 






-i-l 






(0 



•"S- o 


o 


n 


m 


in 


o 


03 


CN • 


• 


• 


• 


• 


• 


• 


in o 


CN 


en 


tn 


n 


a> 


n 


^o 


n 




rH 


,-i 






<Tl r-i 














in 














in 















O 



O 



a\ o 


rr 


m 


n 


O 


n 


ro . 


• 


• 


• 


• 


• 


00 o 


n 


fM 


o 


<Ti 


r^ 


^o 


n 


nH 


iH 






VD rH 












CM 












cn 












^ 












(N 












O 












in 













vo 



n 


CN 


• 


• 


VD 


n 



CM 



4J 

i-l 

to 

•r-l 
O 

w 

r-l 

(0 

o 
s 





>-l O CM o in iH 


O tH O 


M r-l 




ro • • 


• • • 


• 


• • 


• 


• 




^ O "a" c^ 


>! o en 


r- VD VD 


ro CM 






•^o ro r- 


H iH 




<-{ 








n .-1 














.-( 














ro 
















^ 














vo 














in 














in 
















1 


















o 


















-^ 


>i 


to 
















^ 














fO 


I— 1 


-u 


iH 










u 


8 


r-l 


to 










ft. 


to 


-1-1 








J. 


J 


0) 


•-H 


o 








I 


^ >i 

-H d 


& 


pec 


K 








T 


H -H C 


- OI 


en 








» 


* § "c 


H 

J TD 


w : 


r .. 








u 


■J fc. - 


8 S 


iH C 


u to 










' -D 1 


to C 

o % 


71 O 
J -r^ 










L> 1 C 5 


: w w 


•M : 


3 Oi 


>i 








J j-> to 


u u 


-O I 


y 


u 








g c 


H -1-1 -H 




4J 








= 0) iH r 


a u u 


£ r- 


-1 w 


to 






•r 


HO to C 


z -u -u 


f 





-rH 




tH 




J M l-l V 


J to OJ 


U V 


-1 u 


jr i 


-1 


(T3 




n Q) Q) c 


U -r-l JJ 


<D C 


L) <U 


o c 


u 


4J 




Ll CU C 4. 


J 13 W 


x: : 


: j= 


>. s 


— 


e 




u c 


r Qj XI 


-u c 


; -u 


n .1 


J 




U t- 


-1 fti o 


O t. 


3 C 


3 


04 C 


3 



w 
cu 
-u 
to 
J-) 
en 

'S 

-rH 

c 
tn 

D 

o 

c 

-H 

e 

0) 



0) 

x: 



c 

•r-l 

w 

c 
to 

10 



'U 



CO 



•r-l 
C 



u 
o 

to 
to 

w 

0) 

a 

r-l 
O 

c 



277 



Table 8. Percent physician visits for specified race by principal diagnoses and 

ICDA code: United States, 1979 



Principal diagnosis 
and ICDA codes 



Total U.S. 



White 



Asian/ 

Pacific 

Americans 



All classes 

Estimate of visits- 
Percent 



Infective & Parasitic Diseases- 
Neoplasms 

Endocrine, nutritional, and 
metabolic diseases 



Mental disorders 

Diseases of the nervous 
system and sense organs- 



Diseases of the circulatory 
system 

Diseases of the respiratory 
system 

Diseases of the digestive 
system 

Diseases of the genitourinary 
systan 

Diseases of skin and sub- 
cutaneous tissue 



Diseases of the musculoskeletal 
system 

SiTirotcms and ill-defined 

conditions 



Accidents, poisoning, and 
violence 



Special conditions & examinations 
without sickness 



Other diagnoses 
None or unknown- 



556,313,431 
100.0 

3.5 

2.6 

4.1 
4.4 

9.1 

8.9 

13.2 

4.4 

6.6 

5.2 

6.7 

3.1 

9.3 

15.8 
1.5 
1.6 



502,926,839 
100.0 

3.5 

2.7 

4.0 
4.5 

9.4 

8.8 

13.1 
4.5 



5,559,524 
100.0 

4.1 

1.4 

3.2 
1.9 

9.1 

7.3 

15.1 
6.6 



6.3 


5.0 


5.3 


10.2 


6.5 


4.0 


3.1 


1.8 


9.3 


5.3 


L5.9 


23.3 


1.5 


0.6 


1.6 


1.1 



includes data for Indians and Alaskans in the coterminous United States. 

Includes diseases of the blood and blood-forming organs, complications of pregnancy, childbirth 
and the puerperium, congenital anomalies, certain causes of perinatal morbidity and mortality, 
blank diagnosis; nonoodable diagnosis; and illegib.^e diagnosis. 



278 







CM 




r-4 


p-l 




to 


^^ 




JJ 


II 




o 


z 




H 


Vw^ 


03 






U 






n) 


0) 


,— V 


0) 


t-J 


o 


>% 


CO 


r^ 




E 


II 


<J- 


0) 


2 


vD 


t-^ 


^-^ 


in 






O- 




CSI 




IP 


o 




1—1 


II 




to 


Z 




s 


^»^ 



(35 



era 

u 

c 

•H 
f-H 
O 



to 

(U 

X 

c 

3 

o 

XJ 

(0 

c 

-1-1 

o 



c 
o 
u 



CO 

Q) 

XJ 

CO 
Di 

<U 
tj 

c 

0) 

I— I 
CO 

> 

1-1 



CO 

0) 

a: 



V CO 

<-i u 

XI C 

CO QJ 

H S 



CD 

a. 

D 
O 

u 

M 

0) 
00 

< 



CO 

(-1 

CO 
01 

>^ 

<3- 

-J 

I 

u~i 



CO 
u 

CO 
(U 
>^ 

<J- 

CN 

I 

(X5 



CO 


CN 


i-l 


m 


o 


1— 1 


H 


II 




z 




V 


0) 


^-s 


1-1 


CM 


CO 


ON 


P. 


II 


<u 


z 





o 


01 


<J- 


r-H 


II 


CO 


z 


S 


^ 




/—^ 


1—1 


00 


CO 


r- 


J-i 


II 





Z 


H 


^ 


0) 


•-*^ 


r— 4 


m 


CO 


<f 


B 


M 


Q) 


Z 


bn 


N— ^ 




^_^ 




in 


a> 


m 


1—4 


II 


CO 


Z 


2 


^w• 



<J r-l 



O 

CN 



CN 
CN 



O CN 

m r-l 



0^ 
CN 



c» 



ON 



c» 



1—1 CO 



CSI 



in 



vO 



in CN 

1—1 00 



en 






1-1 vD 

r— ( cy\ 



o 
m 



O 
m 



in 

CN 



o 
in 



r^ CO 



00 
CO 









en 
CN 






CSI 



O 
O 



Csl 



)-l 


)-i 


Vj 


(U 


a; 


0) 


OJ 


•o 


T3 


-a 


•a 


o 


^ 


u 


t-i 


CO 


o 


o 


o 


•H 


CO 


CO 


CO 


a. 


•1-1 


•H 


• i-t 


u 


a 


O 


Q 


(U 


>s 


c 


o 


> 


XJ 


o 


• 1-1 


•i-t 


Q) 


• 1-1 


c 


CO 


•1-1 


XJ 


CO 


CO 


g 


CO 


a. 


0) 


< 


•H 




a. 




XJ 




o; 




CO 




Q 




E 








o 




U 




CO 




o 
•■-1 

CO 
S 



OC 



o 
-3- 



1-1 o 

in in 



in 



O 

CO 






00 



o 

>3^ 






in 



o 

CO 



o 



-J- 



<3- 



1—1 

en 



in 



c 


C 


o 


o 


•rA 


• H 


CO 


XJ 


CO 


CO 


<u 


N 


u 


•H 


a. 


^^ 


0) 


CO 


Q 


u 




o 


Q 


E 


1 


0) 


CO 


o 


[d 




U 





c 

O 

a 

XJ 

3 

o 

C! 
I 
CO 

w 
o 

(U 

•r: 
H 



CO 

0) 
M 

CO 
XJ 

C 
O) 

u 

Vj 
0) 
CL 

3 

CO 
)-• 

01 

u 
to 

CO 
Q) 
60 
CO 

XJ 

C 
0) 

c; 
u 

OJ 



4) 




e 




•f4 


• 


XJ 


CN 


OJ 


• 


iw 


1—1 


• H 




1—1 


CO 




• 1-1 



0) 

1-1 

CO 



CO 


O 


0) 


D. 


CO 


XJ 


o 


3 


C 


CJ 


60 




CO 


C 


• H 


o 


-o 


•1-1 




XJ 


M 


CO 


M 


N 


M 


■H 


1 


1—1 


t: 


CO 


en 


u 


O 


o 




s 


CO 


(U 


i-i 


o 


o 






01 


(U 


j: 


j: 


H 


H 





O CO 
Z •'^ 



279 



REFERENCES 



Bureau of the Census 

1981 1980 Census Population Totals for Racial and Spanish Origin Groups 
in U.S. Announced by Census Bureau . Washington, D. C: United 
States Department of Commerce, Public Information Office, Feb 23. 



Campbell, T. and B.Chang 

1973 "Health care of the Chinese 



in America." Nursing Outlook 21: 245. 



Chan, C. W. and J. K.Chang 

1976 "The role of Chinese medicine in New York City's Chinatown." 
American Journal of Chinese Medicine 4: 129-146. 

Chien, C. P. and Joe Yaraamoto 

1981 "Effective psychotherapy for the Asian American and Pacific 

Islander patients." In F. Acosta et al. (eds.). Psychotherapy for 
Low Income and Minority Patients . 



Dohrenwend, Dohrenwend, Bruce P. and Barbara S. Dohrenwend 

1969 Social Status and Psychological Disorder; A Causal Inquiry . 
John Wiley. 



N.Y. 



Drury, Thomas F., Claudia Moy and Gail Poe 

1980 "Going beyong interviewer observations of race in the National 
Health Interview Survey." Paper presented at the American 
Statistical Association, Social Statistics Section, Houston, Texas. 

Feldman, J. J. 

1960 "The household interview survey as a technique for the collection 
of morbidity data." Journal of Chronic Diseases 9:535-557. 

Pinker, A. L. and H. Nisselson 

1978 "Some statistical problems associated with continuing cross-section 
surveys." Pp. 45-68 in N. K. Namboodiri (ed.). Survey Sampling and 
Measurement. N.Y. : Academic Press. 



Frerichs, R. R., C. S. Aneshensel, and V. A. Clark 

1981 "Prevalence of depression in Los Angeles County. 
of Epidemiology 1113: 691-699. 



American Journal 



Gould-Martin, K. 

1978 "Hot, cold, clean, poison, and dirt: Chinese folk medical 
categories." Social Science and Medicine 12:39-46. 

Hurh, Won Moo and Kwang Chung Kim 

1982 "Methodological problems in the study of Korean immigrants: 
conceptual, interactional, sampling and interviewer training 
difficulties." Pp. 61-80 in William T. Liu (ed.). Methodological 
Problems in Minority Research . Chicago: Pacific/Asian American 
Mental Health Research Center. 



280 



Keely, Charles B. 

1971 "Effects of the Immigration Act of 1965 on selected population 
characteristics of immigrants to the United States." Demography 8, 
2 (May): 157-169. 

Kuo, W. 

1984 "Prevalence of depression among Asian-Americans. " Journal of 
Nervous and Mental Disease 172, 8: 449-457. 

Li, F. , N. Y. Schlief, C. J. Chang, and A. C. Gaw 

1972 "Health care of the Chinese community in Boston." American J . 
Public Health 62: 536-539. 

Lieu, Geoffrey, et al. 

1976 Identification of problems in access to health services and health 
careers for Asian Americans . Annandale, VA: JWK International 
Corporation. 

Lin, Tsung Yi and M. C. Lin 

1978 "Service delivery issues in Asian-North American communities." 
American Journal of Psychiatry 135: 454-456. 

Ling, S., J. King, and V. Leung 

1975 "Diet, growth, and cultural food habits in Chinese-American 
infants." American Journal of Chinese Medicine 3: 125-132. 

Liu, William T. (Ed.) 

1982 Methodological Problems in Minority Research . Chicago: 
Pacific/Asian American Mental Health Research Center. 

Liu, William T. and Elena Yu 
In press "Ethnicity and Mental Health." Chapter especially written for 

Joane Moore and Lionel Maldonaldo (Eds.), Urban Ethnics . Beverly 
Hills: Sage Publications, in press. 

Manis, J. G. , M. J. Brawer, C. L. Hunt, and L. C. Kercher 

1964 "Estimating the prevalence of mental illness." American 
Sociological Review 29: 84-89. 

McKenzie, J. L, and N. J. Chrisman 

1977 "Healing herbs, gods, and magic: Folk health beliefs among 
Filipino-Americans." Nursing Outlook 25:326-329. 

National Center for Health Statistics 

1975 Health Interview Survey Procedure, 1957-1974 . Vital and Health 

Statistics, Series 1, no. 11. DHEW Publication No. (HRA) 75-1311. 

National Center for Health Statistics 

1978 Current Estimates from the Health* Interview Survey: United States, 
1977. Vital and Health Statistics Series 10, no. 126. DHEW 
Publication no. (PHS) 78-1554. Hyattsville, Md.: Public Health 
Service. 



281 



Phillips, D. C. 

1966 "The 'true prevalence' of mental illness in a New England State." 
Coininunity Mental Health Journal 2, 1 (Spring): 35-40. 

Pillsbury, Barbara L. K. 

1978 "'Doing the month': Confinement and convaslescence of Chinese women 
after childbirth." Social Science and Medicine 12: 11-22. 

Popov, I.M., and W. J. Goldway 

1973 "A review of the properties and clinical effects of ginseng." 
American Journal of Chinese Medicine 1:263-270. 

Radloff, L. S. 

1977 "The CES-D Scale: A self-report depression scale for research in 
the general population." Applied Psychological Measurement 1: 
385-401. 

Regier, D. A. et al. 

1984 "The NIMH epidemiologic catchment area program." Archives o 
General Psychiatry 41, 10 (October): 934-941. 

Rice, Dorothy P., Thomas F. Drury and Robert H. Mugge 

1980 "Household health interviews and minority health: the NCHS 
perspective." Medical Care 18 (3): 327-335. 

Riley, Matilda White 

1973 "Aging and cohort succession: interpretations and 

misinterpretations." Public Opinion Quarterly 37: 35-49. 

Robins, L. N. et al. 

1984 "Lifetime prevalence of specific psychiatric disorders in three 

sites." Archives of General Psychiatry 41, 10 (October): 949-958. 

Salber, E. J. and A.G. Beza 

1980 "The health interview survey and minority health." Medical Care 18 
(3): 319-326. 

Schaie, K. Warner 

1965 "A general model for the study of developmental problems." 
Psychological Bulletin 64: 92-107. 

Sich, D. 

1981 "Traditional concepts and customs on pregnancy, birth and post 
partum period in rural Korea." Social Science and Medicine 15B(1): 
65-69. 

Suchman, E. A. 

1967 "The survey method applied to public health and medicine." Pp. 
423-519 in Charles Y. Clock (ed.). Survey Research in the Social 
Sciences . N.Y. : Russell Sage Foundation. 

Sue, S. 

1977 "Community mental health services to minority groups. Some 

optimism, some pessimism." American Psychologist 32: 616-624. 



282 



True, Reiko-Homma 

1980 "Health-care issues for Asian Americans." Pp. 365-75 in Michael 

Jospe, Joseph Nieberding, and Barry D. Cohen (eds.). Psychological 
Factors in Health Care. Lexington, Ma: Lexington Books. 



Tseng, W. 
1974 



S., J. F. McDermott, Jr., and T. 
People and Cultures in Hawaii; 



W. Maretzki 

An Introduction for Mental Health 



Workers. Honolulu: Department of Psychiatry, University of Hawaii 



School of Medicine. 



Tung, Tran Minh 

1980 "The mental health of Vietnamese refugees." Unpublished 
manuscripts of the Vietnamese Refugee Research Project. 

Weaver, J. L. 

1976 National Health Policy and the Underserved: Ethnic Minorities, 
Women and the Elderly . St. Louis: C. V. Mosby. 

Yeh, Eng-Kung et al. 

1979 "Psychiatric implications of cross-cultural education: Chinese 
students in the U.S.A." Acta Psychologica Taiwanica 21: 1-26. 



Yu, Elena 
1982 



"The low mortality rates of Chinese infants: Some plausible 
explanatory factors." Social Science and Medicine 16: 253-265, 



Yu, Elena S. H. 

1985 "Studying Vietnamese refugees: Methodological lessons in 

transcultural research." Chapter especially written for Tom Owan 
and Tuan Nguyen (eds.). Southeast Asian Mental Health; A Focus on 
Treatment, Services, Training, Research, Prevention, and the 
Federal Perspective . Washington, D. C. ; U. S. Government Printing 
Office, in press. 



Yu, Elena and Beulah Cypress 

1982 "Visits to physicians by Asian/Pacific Americans. 
20, 8 (August); 809-820. 



Medical Care 



Yu, Elena, Ching-fu Chang, William T. Liu, and Stephen H. Kan 

1984 "Asian-White mortality differentials: Are there excess deaths?" A 
special report prepared for the Department of Health and Human 
Services Task Force on Black and Minority Health, National 
Institutes of Health. 

Yu, Elena, Thomas F. Drury, and William T. Liu 

1981 "Using National Health Interview Survey data in secondary analyses 
of health characteristics of Asian/Pacific Americans: Problems and 
prospects." Paper presented at the American Statistical 
Association Meetings, Detroit, Michigan, August. 



283 



Minority Access to Health 
Care in the Mid-1980's 

Report of the Working Group on 
Health Care Utilization and Financing 




This report was prepared by Caroline Taplin and Ronald H. Carlson, 
Office of Planning and Evaluation, Health Resources and Services 
Administration. 



286 



Minority Access to Health Care in the Mid-1980's 



I. Introduction 

The 1980's have brought renewed interest in the issue of access to 
health care services. The issue is so completely intertwined with the 
issue of cost containment that it is impossible to separate the two 
completely. Rapid changes in the health care environment and efforts to 
contain health care costs have important and as yet not completely 
understood consequences for access to health care at the community, State 
and national levels. The interrelationship of the two issues is such that 
Stuart Altman of the Prospective Payment Commission, speaking to the Health 
Resources and Services Administration early in 1985 identified the 
access-oriented question of uncompensated care as the Achilles' heel which 
could threaten the viability of Medicare's DRG-based reimbursement system 
which has been the centerpiece of Federal health cost containment efforts 
and which has served as a model for several States. 

Measuring levels of access to health care has always been problemmatic 
because of the concept's multi-dimensional nature which demands 
consideration of a variety of indicators and the lack of institutional 
entities with responsibility for maintaining data on a full range of 
elements. The current turbulence within the health care system compounds 
these problems. One focus of the recent reconsideration of access to 
health services has been to take a population-based approach, an approach 
often hampered by data constraints. This paper explores some aspects of 
minorities' access to health care services within this context of system 
change and renewed interest. 

The past two decades have seen significant changes in the nation's 
supply and distribution of health professionals and more frequent usage of 
health services by the elderly, the poor and by minority populations. 
Despite these advances, the Report of the Secretary's Task Force on Black 
and Minority Health has described in detail the disparity in health status 
which continues between minority populations and the White majority . This 
paper will examine, to the extent possible, minority experience around some 
of the indicators which taken together describe access to health care; it 
will note areas where better data are needed and it will raise some of the 
many questions which remain unanswered about the experience of minority 
groups in seeking and receiving health care services. 

This paper draws exclusively on already existing material and relies 
heavily on three sources: 

• Data from the National Health Interview Survey for the years 
1978-1980 

• The 1982 National Survey on Access to Medical Care conducted by Lou 
Harris and Associates for the Robert Wood Johnson Foundation 

• Data compiled by the Secretary's Task Force on Black and Minority 
Health's Subcommittee on Cancer. 



287 



These sources deal extensively with a range of access related 
indicators for Black and Hispanic populations compared to Whites. There is 
a lack of comparable information for many of these indicators for Native 
Americans and Asian Americans which limits the discussion of these groups. 

II. Access: Definition and Approach 

The concept of access to health care is multi-dimensional. It 
describes an individual's willingness, ability and actual entry into the 
health care delivery system. It is influenced by characteristics of the 
delivery system such as the level and distribution of available resources, 
cost, provider characteristics and convenience, and by characteristics of 
the person seeking care such as family income, education, insurance 
coverage and socio-cultural attitudes. This paper attempts to organize 
existing data within a model which suggests the many dimensions of the 
access issue and some of the interconnections among its attributes. It 
draws on a model for access developed by Aday and Andersen which considers 
both probable or potential access as illustrated by population 
characteristics and characteristics of the health care delivery system and 
actual or realized access as indicated through utilization measures and 
degree of patient satisfaction. ( 1) The figure on the following page, 
based on this model, arrays the characteristics of access this paper will 
consider. 

This paper will discuss what the individual and system variables are 
which characterize minority populations and how these populations' 
experience in utilization and satisfaction compares with the White 
population. It differentiates the experiences of minority and White 
populations according to their use of and satisfaction with health 
services. It attempts to shed light on these differences by highlighting 
differences in the characteristics of the individuals and the communities 
in which they reside. 

The paper presents a national overview and as such is subject to the 
vulnerabilities of aggregate analysis. Though they draw on relatively 
large national sample groups, the studies relied on do not contain 
subsamples of minorities large enough to reveal regional or subregional 
patterns. A national average statistic for any indicator tells nothing 
about the range of observations encountered, regional patterns or 
differences experienced within a jurisdiction. In a limited way, however, 
this paper will use sub-national examples as illustrations of the variety 
of experiences encountered within the country and the diversity and 
complexity of the issues associated with access to health care. 



288 



ACCESS TO HEALTH CARE AND ITS ATTRIBUTES 



POTENTIAL ACCESS 



REALIZED ACCESS 



Individual 



Predisposing 



Age 6 or less 
Age 65 or over 



Enabling 

Financing: Income, Employment, 
Type and Extent of 
Insurance Coverage, Out-of-Pocket 
Expense 

Organization: Regular source of 
Care, Type of Care Provider, 
Particular Provider, Travel Time, 
Waiting Time 

Need 

Perceived Health 
Disability Days 
Limits on Activity 
Recent Medical Emergencies 
Serious Illness in Family 

Community 

Availability 

Physician/Population Ratio 

Location 

Region 

Rural Residence 

Central City Residence 



Objective 

Use 

Physician Visits 
Site of Visit 
Preventive Activities 
Dental Visits 
Hospitalization 
Difficulty in Receiving 
Care 

Use Relative to Need 



Subjective 

User Satisfaction with: 

Travel Time 

Waiting Time 

Time with Physician 

Information Received 
Out-of-Pocket Expense 
Quality of 
Overall Visit or Stay 



289 



III. Potential Access 

Potential access indicators measure the characteristics of the 
population under consideration and of the health care delivery system that 
influence whether care will be received. 

Andersen has grouped individual determinants into three classes: 

Predisposing; 

Enabling and 

—Need. (2) 

Predisposing characteristics are those factors which suggest the 
likelihood of using services and which exist in the individual before the 
onset of illness. They include characteristics which cannot be changed 
such as age, sex, race or ethnicity and those which can such as educational 
level, knowledge of good health practices and general health care attitudes 
and beliefs. Enabling factors describe the resources available to a 
person should he decide to seek care both individually and within the 
community. Need attempts to measure health status and immediate causes 
for seeking care. 

Predisposing factors 

Table 1 illustrates information on some predisposing access 
indicators which are demographic in nature. As the table shows, a larger 
percentage of the minority population than the White population is young, 
suggesting a proportionately higher need for preventive services 

The overall level of child bearing for minority women is higher than 
for White women. The fertility rate for Black women is 2.3 children 
compared to 1.7 for Whites. The average number of children in U.S. 
Hispanic families is 2.3. The American Indian birth rate is almost twice 
that of all U.S. races. Asian Americans have larger households and average 
family size than the U.S. population as a whole. (3) 

Smaller percentages of minority populations are elderly. With the 
exception of Asian Americans, minority populations achieve lower levels of 
education than the White population. This has important consequences for 
their income, ability to pay for care, and insurance status. It can also 
affect knowledge of services available and how to purchase services. 

Enabling factors 

Table 2 illustrates variations in income and employment among racial 
and ethnic groups. 



290 



c 

CD 
Q. 
CO 



a 



00 

■ 



o 



C/7 



CJI 



en 



on 



in 



CD 



to 



o 
CO 

r— I 

CD 






00 



1"^ 



CM 



■ 

c^ 



CO 
CTs 



(D 

4-1 






c^ 



LP, 
ON 



Q_ 
O 
Q_ 



* 
CO 
(D 
O 
CD 



LA 



CO 
CO 



u 
o 



(D 
CJi 



(-1 

CD 

> 

o 

(-1 
o 

LA 

VO 

cu 
>a: 



CO 

c 
o 

CO ID 

u en 

.— I (D CD 
O Q- 

O ^-- (4- 

x: o 
(J -a 

4_) C_l f_l 

14- dJ CD (D 

O ^ (D > 
Q. >^ O 

CO E 

C-i O lA tj 

CD CJ CSJ O 
(D 

>- 



CO 

CD 
(U 

>, 

00 



00 



0) 
0) 



o 

C_) 

CO (U 

(4 C-l 

CO o 

CD e 

>- 
t-l 

<f o 





CO 








en 






c 






■H 






c 




"O 


t-l 




C 


CD 




CD 


UJ 




-* 


T3 




<f 


•a c 






C CD 






CD 




• 


4-) 




O 


CD C 




z 


00 (D 


\ 




fA e 


^ 


r. 


>^ 




CD 


• o 




MD 


O —i 






Z Q. 




Q- 


E 






^Ld 




CO 


O 


ID 


CSI CO 


■H 


1 O 


t-l 


Q. -H 


HI 


4J 


to 


■a CO 




C -H 


■a 


CD 4-> 


c 


CD 


CD 


.r.4J 




csi tn 


00 


<J- 


VO 


^ u 


rA 


o 




• JD 


• 


O CO 


O 


Z _J 


z 


-Cl- 


^ 


o o 


a 


^ 


Cvj 


1 D 


1 


£2- CD 


Q- 


(D 




CO (-1 


CO 


(U 3 


CD 


■H CD 


•H 


U 


t4 


QJ • 


OJ 


to to 


to 


-ID 


^ 


CO 




CO 




-t-J 


>^ 


4-) 




C-l 


-Q 


c-l 




o 




o 




Q. 


■o 


CL 




(D 


(U 


(D 




EC 


CO 


ce 




C 


■H 


c 




o 


1— 1 


o 




•H 


-Q 


■H 




-P 


3 


4.J 




CO 


Q. 


CD 




r-H 




r— 1 




D 





D 




Cl 


t-l 


Cl 




o 


CO 


o 




Q_ 


CD 


Q- 




-u 


-u 


4-1 




c 


CD 


C 




(D 


■D 


ID 




f-i 




t4 




t-l 


(D 


tH 




a 


O 


D 




OJ 


t4 


C-J 




o 




-c<- 


». 


co 


CO 


3 t-l 


3 


CO o 


CO 


C -Q 


c 


CD CO 


CD 


C-3 _l 


C-) 


(D 


CD 


-C • 


-C • 


4-> CO 


4-1 CO 


4-> 


4-1 


c<- CD 


I*- CD 


O -D . 


O -D 


<r 




3 -O 00 


D -o 


CD (D On 


CD (D 


-C r- 


(D .C 


tn CO 


t4 CO 


13 -H >^ 


D -H 


CD •-< t-l 


CD ^ 


-Q CD 


-Q 


• 3 a 


• D 


to Q. C 


to Q- 


• C CD 


• C 


ZD 


3 


r-) 


ID 


D 



CJ 



o 
to 



291 



a 



CO 

o 

I— 
< 

CJ 






<: 

CO 
•a: 






CJ 



^^ 



25 
ri! CJ 

^^ 

Q_ 



O 
ON 



cn 

c 

CO 

c o 

CD -H 
H t-i 

CO <u 

a: e 

< 



00 



ON 



NO 



CTv 



CSJ 



c 

CD cn 

o c 

H CD 

C-i -H 

CD "O 

E C 

<a! >'^ 



csj 

LA 



lA 

c-g 






CM 



co 

CO 



O 



(D 

■a: 



rA 
cn 



CM 

NO 



(D 

> 
o 

u 

O 

lA 
NO 

(D 

cn 
<: 





CO 








c 








o 








CO 


(D 






fH 


Di 




1 — 1 


(U 


CD 




o 


Q- 






o 




C_ 




r: 




O 




o 


■D 






CO 


CD 


CO 






-tJ 


u 


t-i 


U- 


CD 


CD 


(D 


o 


r-H 


(U 


> 




Q. 


>. 


O 


CO 


E 






t-1 


O 


lA 


(-1 


CD 


CJ 


(Nl 


o 


CD 








>- 









CO 
C-1 
CD 
(D 



Ol 




1— 1 




O 




CJ 




cn 


(D 


tl 


tH 


CD 


O 


(D 


E 


-^^ 


u 


<t 


o 



o 

CD 
C_) 
Q_ 



(D 

4-) 

CL 
CD 



O 



c 
o 



Q. 
O 

Q_ 

<D 

SI 



CO 
CO 

c 

(D 
CJ 

CD 
00 

(JN 



CO 
CO 

c 

CD 
CJ 

(D 



3 
CD 
(D 

CD 



CJ 



o 
cn 



292 



o 



U7 



tn 






UJ 



CQ 



CO 
0\ 






CD 



O 



a 


CO 






H 


(^J 


(JN 


CJs 


c 


CM 




• 


CD 


•« 


o 


CVJ 


a 


so 


l^^^ 


SO 


CO 


^- 







o 
cc 

CQ 



VO 



«> 



ON 


MD 


ON 


• 


lA 


LH 


•^ 


r^ 



M3 



lA 





r^ 




O 




VO 


CD 




-U 


<r 


■H 


CM 


x: 


•y^ 



so 



CO 





r^ 


m 


rA 


(D 


<f 


a 


rt 


CD 


rA 


cc 


CVI 




■t«- 


r-H 




r-H 




<: 


<D 




E 




O 




O 




c 



vo 



tjN 



c 

CD 






s 




o 


(D 


1 — 1 


C 


(U 


•H 


J3 


f— 1 


03 


>^ 


c 


4_> 


o 


l^ 


CO 


CD 


[-1 


> 


(D 


o 


Q. 


Q. 





(-1 






0) 




■D 


> 




C 


o 




CD 


a 




VO 


s 




'" 


o 
■— 1 




CO 


(D 




c 


-D 




o 






CO 


1-1 


CD 


t-l 


(D 


C 


CD 


> 


■H 


□_ 


O 


>— { 



CD 
+-> 
CD 



c 

(D 

e (D 
o o 

i-H E 

Q. 

E U 
0) O 

c 





CO 




cr 




c 




•H 




c 




tJ 




CD 




LjJ 




■D 


•a 


C 


c 


CD 


CD 






■)-> 


00 


c 


CO 


CD 


f^ 


E 




>, 


• 


O 


o 


^ 


z 


CL 




E 


- 


LiJ 



o 

CSJ CO 

I o 

Q_ -H 

■a CO 

C -H 

CD 4-> 

CD 

.^ j_i 

CM tn 
<t 

r— t-i 

O 

• J3 

O CD 



..Ct- 
O O 



M3 

I 

Q. 

CO 

ID 



13 
CO 
CD 
(h 

•H CD 

(D • 

Ul u-> 

• 

CO 

"O 
(D 

il 
CO 



u 
o 

CL 
(D 
CC 



c 
o 

•H 13 



CD 

r— { 

3 
Q. 
O 
CL 



c 

C-l 
C_) 



3 

Q. 

(D 

[-1 
CD 

CD 

4-1 

CD 
■D 

CD 
O 
U 
O 



CO 

D t-i 

CO o 

C JD 

(D CD 

C_) _l 

CD 

JZ • 

4-> CD 

l4- CD 

O "D • 

<r 

3 ■□ CO 

CD CD (J\ 

(D ^ T- 

tj CO 

3 -H >^ 

CD ■-^ U 
JD CD 

• 3 3 

tn Q. c 

• C CD 
=) D r-) 



C_) 
CC 



c 

CD 



o 

vo 
I 

Q. 

CO 
(D 

•H 
f-l 
(D 

LD 

-o 

C 
CD 

CO 
vo 



O 
CM 



CO 
ID 
■H 
U 
0} 

to 



CO 

u 
o 

CL 
Q) 

CC 

c 
o 

■H 
CD 

r-H 

■D 
Q. 
O 
CL 

4-> 

C 
Q) 
C-l 
C-l 

C_) 



CO 
CO 

c 

CD 
CJ 

CD 

^ • 

4J CD 

4-) 

c<- CO 

O -D 

3 -D 

CO (D 

<D ^ 

fcj CO 

O -H 

CD ^ 
X3 

• 3 

tn CL 

• c 



o 

CJl 



293 





OH 


CO 






o 


z 






la. 








OO 


1 — 1 






q: 


cc 






a 


Ld 






1 — 


s: 






< 


<c 






CJ 








f — 1 


z 






Q 


<: 






Z 


1 — 1 






1 — 1 


en 




•=t 




<c 




CM 


1 — 








Z 


Q 


o 


Ld 


UJ 


Z 


00 


-J 


2: 


<: 


ON 


CD 


>- 






-=C 


o 


on 




h- 


_j 


2 






Q- 


< 






2: 


1 — 1 






Ld 








Q 


1—1 






Z 








<: 


<: 






Ld 


CJ 






2: 


1 — 1 






o 


q: 






CJ 


Ld 





c 
to 
o 

H C 
t-i CO 

e -D 
<: c 



cvi 

CM 



MD 









<t 


LA 


^0 


CVI 


• 


• 


r^ 


r^ 


00 


»> 


CM 


lA 


r^ 






T— 






■t«- 







c^ 



)-t < 



CO 

O 

CO 



0\ 



■fc«- 



e 
o 
o 

c 



<f 


o 


• 


m 


CM 


CM 


'— 


^ 



lA 
M3 



C 
CD 
•H 
■D 
CD 



2 




O 


(D 


■— 1 


c 





•H 


J3 


>— 1 


CD 


>^ 


C 


-P 


o 


C-l 


CO 


0) 


^ 


> 


(U 


o 


Q- 


Q. 





tn 






OJ 




■D 


> 




c 


o 




CO 


Q. 




MD 


S 




'" 


o 

1— 1 




CO 


QJ 




c 


J3 




o 






CO 


Ca 


<D 


(-1 


(D 


C 


(D 


> 


•H 


Q- 


O 


1— 1 



-U 
CO 
Qi 

+J 

C 

e (u 
>> u 

o o 

^ E 
Q. 

B U 
0) o 

c 



CJ 



c:: 

00 
CJ 

a_ 



CJ 

tj 

(D 
-U 

Q. 

CD 
JZ 
CJ 



o 



c 
o 



Q. 
O 
Q- 

0) 



CO 

CO 

c 

(D 
CJ 

CO 



CO 

CO 

c 

CD 
CJ 

(D 

4J 

14- 
O 

CO 

(D 

C^ 

:j 

CD 



tn 



CJ 

ct: 



o 

LD 



294 



Black, Hispanic, and American Indian families have substantially 
lower incomes than the White population. Black family income is only 55 
percent of White family income, while Hispanic family income is 66 percent 
of the majority group. Within the Hispanic population income varies with 
Puerto Ricans reporting the lowest family income. (4) 

Median family income for American Indians is also low. Relatively 
high income levels are found in the Asian American population, many of whom 
live in households containing several working adults. 

Minority groups, with the exception of Asian Americans, also 
experience higher unemployment rates, with the consequence that fewer 
minorities have access to employment-based third-party health insurance the 
prevailing national mode for the financing of health care. Lower incomes 
result in less disposable income available for the direct or indirect 
purchase of health care. 

A prerequisite for access to care is a method of paying for services. 
This usually takes the form of some type of third-party insurance, either 
public or private. The Robert Wood Johnson survey found that in 1982 
approximately 9 percent of the American adult population under 65 reported 
no health insurance coverage of any kind. This percentage had held 
relatively constant since 1976.(5) In 1983 the Census Bureau looking at 
all ages of the American population found 15.2 percent of all persons to be 
not covered by either private or public health insurance. Fourteen percent 
of Whites lacked health insurance, 21.8 percent of Blacks and 29.1 percent 
of Spanish origin persons. (6) 

Table 3 illustrates insurance coverage by type for White, Black and 
Hispanic populations under 65 years of age. Minority populations are two 
to three times as likely to be uninsured as the White population. 
Inability to pay is the most commonly cited reason for not having health 
insurance. (7) 

Minority populations are also more likely to be covered by Medicaid 
than the White population. The especially high Medicaid coverage rates 
for Blacks and Puerto Ricans can probably be attributed to two factors. 
Greater percentages of these families are headed by single women than other 
population groups, thereby increasing the likelihood of categorical 
Medicaid coverage through Aid to Families with Dependent Children. Also, 
it is probable that many of these population groups live in states, such as 
New York, with relatively generous optional Medicaid coverage. (8) 

Although Medicaid has provided health insurance coverage to many 
minority families, not all States offer the full range of services allowed 
under the program and eligibility requirements vary from State to State. 



295 



c 

CO 
-Q 

13 



CM 



CM 









03 



c 

CD 
Q. 
CO 



I 

c 
o 



UJ 

<c 

u. 
o 

LO 

<c 

UJ 

>- 

lA 

cc 

Ld 
Q 



to 

CJ 



< 

Q. 






CD 



en 



Ld 






o 

C_) 



UJ 
CJ 



< 

an 



in 



c 

CD 
O 
•H 

q: 
o 

(D 



C 
CO 

o 

•H 
X 
ID 






CJN 

i<A 



CJN 



o 


r~- 


ON 


• 


• 


• 


(J\ 


o 


C3N 


lA 




CVJ 



o 

■H 
C 
CD 
Cl 
CO 



VO 



lA 



PA 



lA 
CSJ 



PA 
CM 



O 
CD 

CQ 



PA 


O 


CO 


• 


• 


• 


T— 


o 


r- 


VO 


CM 





tu 

■H 



T- 


t~- 


1^ 


lA 


• 


• 


• 


• 


VO 


CM 


CO 


CM 


00 









■D 

C 
CD 

O 
■H 

C 
CO 

a. 
m 



on 

c 
o 



CO 



c 

CD 



CO 
>> 

0. 
■D 

c 
CO 

(U 
CJl 
CD 
U 

> 

o 

C_) 
(U 

o 

c 

CD 

:3 

CD 

C 



CD 
(D 



CO 
O 



•a 

c 

CD 



OD 












CJl 












CO lA 












C-i \o 












(D 












> U 












O 0) 












CJ T) 




>, 








c 




r-l 








(U =) 




c 


■D 




^^ 


o 




o 


CU 




o 


C CO 






t-i 




CO 


CO c 




T3 


0) 




On 


tj o 


CD 


•H 


> 






13 CO 


-(-> 


CO 


o 




1 


CO u 


CO 


o 


CJ 


tj 


O 


c a> 


> 


•H 




0) 


r^ 


l-H Q_ 


•H 


T3 


4-> 


-C 


ON 




t-i 


(D 


o 


J-1 




o? 


Q. 


s: 


2 


o 


^^ 







o 

c 

■H 
> Q. 
(U O 

1— Q_ 



CJ 



C3 
CO 



296 



Mexican Americans are the population subgroup least likely to report 
health insurance. They are also the group least likely to report 
unemployment as a reason for non-coverage. This suggests the likelihood 
that members of this group are often employed in organizations that do not 
provide health insurance. (9) 

1977 data show that employment is no guarantee of health insurance. 
Across population groups, 22 percent of the working poor lacked any form of 
health insurance throughout the year. (10) Nearly one fourth of all 
agricultural workers are uninsured during part of the year, with 16 percent 
uninsured for the entire year. Insurance coverage also is low among blue 
collar and service workers. (11) 

The Robert Wood Johnson Foundation survey found that during 1982 
about 21 percent of Americans experienced changes in their insurance 
coverage. On balance, Americans experienced a net positive change, but 
the poor, and especially poor minorities were more apt to have their 
coverage dropped or reduced. (12) 

The 1977 National Medical Care Expenditure Survey (NMCES) examined 
continuity of insurance coverage, finding 84 percent of its respondents 
always having insurance coverage, 7 percent insured part of the time and 
9 percent always uninsured. For both Whites and Blacks, continuity of 
insurance is directly related to income. For Hispanics the relationship 
between continuity of insurance coverage and income is less clear, but 
generally tends towards an association between the two factors (Table 4). 
This same survey found the largest number of ambulatory visits (4.0) among 
the always insured and the lowest number (2.2) among the always uninsured. 
Whites have higher numbers of ambulatory visits at nearly every income 
level. (13) 

Nearly all the U.S. population over age 65 is covered by Medicare. 
Because Medicare does not provide full insurance coverage for all health 
needs, in many cases beneficiaries supplement their coverage with private 
insurance. Overall 65.2 percent of the over 65 population have both 
Medicare and private insurance coverage, 20.4 percent have Medicare only 
and 10.6 percent have both Medicare and Medicaid. The White population 
supplements Medicare with private insurance more than twice as frequently 
as Blacks (69 percent versus 31 percent) .(14) 

The 1977 National Medical Care Expenditure Survey collected 
information on out-of-pocket expenditures for personal health services. 
Expenses per person for those persons with out-of-pocket expenses were 
relatively constant across ethnic/racial lines: $196 per capita for 
Whites, $180 for Blacks and $191 for Hispanics, although the lower incomes 
for minority groups makes the burden of these expenditures greater. 
Blacks and Hispanics were considerably less likely than Whites to have high 
annual out-of-pocket expenses and while 3.3 percent of all White families 



297 





Q 






2 






<C 






^ 






CD 






=D 






1 — 






< 












CO 


^ — s 




UJ 


^s 




u 






_J ^ 




ID 


INCOME 
States 




1 — 






_i 
< 

UJ 


AND 
ited 




>- 

CD 


ID 


< 

1 — 


CO 

1 — 1 
CO 
1 — 1 

> 


BACKGRO 
d data: 




>- 

a 

1 — 
<: 

_i 

2: 


[C/RACIAL 
househol 




u. 






o 


^ s 



OD 



< 
UJ 



UJ 

q: 

CO 



CD 



o 
en 

Q 

UJ 

q: 

ID 
CO 



Q 
UJ 

cc 

ID 
IS) 



>- 



en 




CJ 




1 — 1 




1— 




CO 


en 


l-H 


c 


cc 


o 


UJ 


M 


1- 


tH 


CJ 


(U 


■=n 


CL 


ce 




■a: 


.—1 



CJ 



CN 









-^d- OD ur\ o 
r^ CM csj r^ 



^ OS r^ <t 
• • • • 

<t r^ -^ r^ 



CD vo T- r^ 

• • • • 

<f O- <3- -^ 





Q) 








E 








O 








o 


OJ 






c 


E 






■H 


O 


0) 






o 


E 




5 


c 


O 




o 


•H 


o 




■-H 




c 






0) 


•H 




t-i 


■— ( 




u 


(D 


-o 


^ 


o 


x: 


■o 


CJi 


o 


4J 


•H 


•rH 


Q. 


o 


s: 


n: 



00 r^ <a- 

... * 



^ j|^ ^ j|c 



<J- <t r-- 

. . . ?)e 
CM CS| CM 



* * * * 



00 T— r- lTn 
.... 

t^ r^ r^ KN 



t^ LP, <)■ 0\ 
.... 

LA r^ f^ CM 







(D 




















Q 










o 


(D 








c 


E 








•H 


O 

o 


E 






s 


c 


O 






o 


■H 


CJ 






^ 


(D 


c 

•H 






tJ 


1—1 






u 


(D 


•o 


H 


^ 


o 


il 


TD 


Qi 


o 


o 


4-) 


■H 


•H 


CO 


Q- 


CD 


s: 


X 



OD 



C 

CO 

CL 



u 
o 
o 

Q_ 



o 
o 

c 






S ZL. 



N 
•H 

m 



•H 
E 
(IJ 

it. 

bi 


■o 

m 

•n 
■O 



298 



i 



reported no out-of-pocket expenses for medical care, 12 percent of Black 
families and 11.4 percent of Hispanic families reported no such 
expenses. (15) 

Organization 

The ease with which a person enters the health care delivery system 
and what happens to a person upon entry are organizational elements which 
have bearing on one's access to health care. Table 5 illustrates 
variations among some of these factors. Having a usual source of care has 
been found to be a good predictor of utilization and suggests a greater 
continuity of care, which is usually associated with improved health 
outcomes. While there is little variation among population groups in the 
percentage of persons having no regular source of care, Whites are more 
likely to have a particular doctor. Blacks are disproportionately likely 
to rely on a hospital outpatient department or emergency room for care, 
health care sources which are not conducive to continuity of care. (16) 

Having some kind of insurance coverage is related to whether or not 
people report having a usual source of care. One-fourth of the medically 
uninsured report having no usual source of care. (17) 

The convenience of health care sources can be gaged in part through 
consideration of waiting and travel time. Blacks and Hispanics both report 
longer office waiting times than Whites, with approximately twice as many 
Hispanics reporting waiting times in excess of 30 minutes as Whites. The 
1977 National Medical Care Expenditure Survey examined travel times to 
usual sources of medical care. It found that about four-fifths of the 
population had to travel less than 30 minutes to its usual source of care. 
While Blacks and Hispanics were more likely to spend more than 30 minutes 
in travel time, 75 percent of Blacks and 76 percent of Hispanics were 
within 30 minutes of their usual source of care. (18) 

There has been rapid and continuous growth in the number of health 
maintenance organizations (HMO's) in the recent past. Since 1973, when the 
National HMO Act was established to spur the growth of HMO's, membership 
has increased from 4 million to over 15 million. In 1984, 9 percent of 
privately insured households nationwide reported at least one family member 
who belonged to an HMO. A 1984 Lou Harris survey found HMO members to be 
more satisfied with their access to 24 hour access to doctors and emergency 
care than non-members. Few data exist on minority participation in HMO's, 
but this same survey found that in 1984 for households whose head has 
health insurance coverage other than Medicaid or Medicare, 9 percent of 
Whites and 10 percent of Blacks reported HMO membership. ( 19) Changes in 
the Medicare and Medicaid programs now encourage expanded use of HMOs by 
program beneficiaries. 



299 



Q. 



csi 






CO 

o 

c 

CD 
CL 
CO 



lA 



CD 






CJ 

o 



ce 
o 
I — 

CJ 

< 



CD 



u 
<: 

CD 






CSI 



est 
00 
C?\ 






o 

CN 






CO 



C7I 

c 
o 

E 

<: 

(D 

u 

CD 
C_) 

^ 

1-H 
CO 
CD 

o 

CO 
CO 

CD 

o 
o 

=j: 

en 

c 



CD 
Q. 











C 












cu 












B 












4-> 






4-> 






(H (D 






3 




m 


CD (-1 






CD 




^ 


Q. CD 










CD 


OJ C_) 


CO 




0) 




C_) 


Q 


(D 




h 






t(- 


4-1 




CD 




U- 


4-1 O 


D 




CJ 




o 


C 
0) Q) 


C 


(^ 


14- 




(U 


•H U 


s: 


O 


O 




o 


4-> tj 




+J 




f-l 


u 


CD D 


a 


o 


(D 


o 


3 


Q. O 


r^ 


o 


O 


-t-> 


o 


1 to 




Q 


i:^ 


o 


U1 


4-> 


(4 




3 


o 




3 t. 


(U 


u 


O 


Q 


u 


O CD 


> 


CO 


CO 




CO 


i-H 


o 


rH 




C-i 


i-H 


-H 3 




3 


(-1 


CO 


3 


CO cji 


CD 


O 


CO 


1— 1 


en 


4-1 (D 


E 


•H 


■—1 


3 


QJ 


•H cn 


■H 


-U 


ZJ 


o 


Dl 


Q. 


t— 


[-1 


en 


■H 




CO CO 




CD 


(U 


4-1 


O 


o CO 


Dl 


□. 


a: 


f-l 


Z 


n: 


C 






CD 




cc 


•H 


JZ 


JZ 


Q. 


x: 


jr uj 


4-> 


-p 


-p 




-p 


4-1 


•H 


■H 


■H 


O 


■H 


•H (J 


CD 


s 


s 


2 


3 


s o 


3 



CD 

■a 

c 

CO 



o 

CO 
•H 
CD 



c 

(U 
CO 

CD 

■o 

c 
< 



bJ 

CJ 

r: 
o 
in 



300 



Community Factors 

A person's access to care is dependent on the availability of health 
resources, both health care professionals and institutions willing to serve 
him within his community. The number of physicians in the nation has grown 
from 291,000 in 1970 to 438,000 in 1982.(20) The ratio of physicians to 
100,000 population has increased from 142.7 to 192.2 in the same 
period. (21) The growth of the physician pool has encouraged the dispersion 
of doctors to previously underserved areas so that specialty care is now 
increasingly available in smaller communities. This dispersion, however, 
has not been even and many rural areas and inner cities remain unable to 
attract and retain an adequate complement of physicians. 

The availability of care implied by the increased numbers of 
physicians may be misleading. Measures of physician supply must consider 
more than simple population-to-provider ratios. The location of the doctor 
within the area, the nature of the physician's practice and one's 
willingness to accept Medicaid or Medicare patients affect the true 
availability of care, especially to poor individuals. The composition of 
the physician pool in American cities is changing and while increased 
numbers of physicians have entered specialty practice, there is a 
continuing shortfall of primary care providers, especially in inner city 
areas. 

The degree to which health care providers participate in the Medicaid 
program is an important determinant of physician access. Low Medicaid 
reimbursement rates discourage private physicians from accepting poor 
clients. A recent study by the American Academy of Pediatrics showed that 
the number of pediatricians participating in Medicaid in the 13 states 
surveyed dropped three percent between 1978 and 1983.(22) While this 
reduction is not large, it may signal a trend toward reduced access to care 
for the insured poor. Other surveys have shown, in Hartford, Connecticut 
87 percent of the obstetrician-gynecologists in the area refused Medicaid 
patients and one Kentucky county where previously 12 physicians had served 
Medicaid patients found this number reduced to four. (23) 

Table 6 shows the distribution of White, Black and Hispanic 
populations by region of the country and type of community. Blacks and 
Hispanics are more urban than the white population and more likely to live 
in a central city. Thus, the emerging public policy debate over 
uncompensated care, a problem whose impact is greatest for urban and public 
hospitals which are major sites of care for the urban poor, both for 
hospitalization and primary care, is of particular importance to minority 
populations. 



301 



UJ 
Q. 

>- 






to 72 
o 



c_) 



^o 






or 

>- 

CD 

UJ 
CJ 

z 

UJ 
Q 
I — I 
LO 
UJ 

a: 
Q 






tn 
o 



CO 
0\ 



c 

•H 
Ol 
■H 

o 

en 

•H 

c 

CO 
Q. 
CO 



o 

CD 

.—I 
CD 



4J 



ON LPl I — lA 



GO r~- f^ ON 



00 o r^ LfN 



CD ifS t^ 
ir\ Ks T- 



00 o NO vo 



CO O C3 o 
r- CSJ LA r- 



LTi \0 <J- lTN 



r- O r- 
LTN CM CNI 



-:3- VO ON 



CSI NO r- 
CSI CNJ t^ 



CJN 



00 ON 1^ o 



LfN O T- CNJ 

CM <t ^N, 



C 

o 



- o 

t^ CO 
'- CO 

• c 

o 

Z -H 
4-> 

^ CO 

O E 
NO U 

1 o 

c 

CO -H 
CD 

■H -a 
(-1 l-l 
0) o 

CD 

•- CO 

en ■J 

4-i O 

u :r. 
o ^- 

CL 
(D 

ce 



c 
o 

■H 

-P 

CO 



o CO 

Q- +-> 



4-> 

c 

(D 

(-1 
C-l 

:3 

CJ 
CO 

CJN 



tn 



E 

ti-- CO 



o 

CO 

en 

















>^ 




E 




•H 
















■p 




u 




4-) 


CM 














•H 




CO 




CO 


00 






CO 








CJ 




u. 




•H 


ON 






ti 














E 


4-> 


V— 






-ij 








1— 1 




c 


(H 


CO 






+j 


c 








CO 







CD 


4-> 


CO 




cn 


CD 








u 


d 


z 


U_ 


CO 


■a 




CO 


CJ 






ID 


4-) 


0) 








1— 1 




(D 








O 


c 


X 


< 


<: 




o 




x: 


H 


X 




c 


(D 


4-1 


CO 


cn 


• • 


^ 


c 


-u 


-U 


-u 


4-) 


CD 


CJ 


o 


2: 


s 


UJ 


CD 


o 


^ 


Ui 


D 


CO 


T) 






cn 


C/l 





cn 


•H 


o 


o 


O 


(D 


■H 


eC 


<: 




1 


cr 


D 


Ol 


2 


z 


cn 


S 


cn 


C/5 


C/1 


c 


c 


rs 


o 


CD 










CD 


2: 


2: 











:r. 


q: 


a? 


s? 


s? 


s? 


cc 


tn 


tD 


2 


z 


in 



CM 

CO 

c^ 



302 



Adequate payment to providers for care delivered to the medically 
indigent, that is, the uninsured and underinsured, has become a critical 
question of national health policy. From the provider perspective, 
uncompensated care represents the debt incurred after services are 
delivered but no or inadequate payment is made. It is the total costs of 
care delivered to: 

• Medically indigent for which there is no public reimbursement. 

• Medically indigent for which there is inadequate public 
reimbursement. 

• Bad debt incurred when patients do not, but presumably can, pay 
their bills. 

Many health care facilities, through receipt of Federal grants, loans 
or loan guarantees, have accepted an obligation to provide a level of 
uncompensated care to qualified persons in need of care. These obligations 
were incurred under Title VI of the Public Health Service Act, also known 
as the Hospital Survey and Construction Act of 1946 (popularly known as the 
Hill Burton Program) and Title XVI of the Public Health Service Act, also 
known as the National Health Planning and Resources Development Act of 
1974. The total number of facilities with uncompensated care obligations 
as of January 1, 1985 was 4,653 of which 2,801 were hospitals. But the 
volume of care provided through this program is not sufficient to provide 
adequate health care for the growing number of medically indigent. (24) 

The problem of uncompensated care affects minority populations 
disproportionately because minority members are more likely to lack health 
insurance coverage and thus are more likely to be unable to pay for care. 
Moreover, urban hospitals and public hospitals bear a disproportionate 
share of the uncompensated care burden. 

Community hospitals provided $7.5 billion in uncompensated care in 
1982. Of the $7.5 billion, $2.3 billion, 31?o was reported as charity care; 
the balance, $5.2 billion, resulted from bad debt. These figures from the 
1982 AHA Annual Summary are based on hospital charges and not the actual 
costs of uncompensated care. (25) 

Analysis of 1981 Hospital Discharge Survey data from the National 
Center for Health Statistics revealed important information about the kinds 
of care received by uncompensated care patients: 

• The cases were most likely to be either maternity or 
accident-related; 

• The most prevalent surgical category for this group was 
obstetrics; 

■ Problems relating to premature birth were common for newborns; 
and 

• The patients were likely to have had surgery on the day of 
admission. (26) 



303 



Medicare has adopted a prospectively based inpatient hospital payment 
methodology that is intended to create incentives for hospitals to 
operate in a more efficient manner. As of December 1984, there are six 
Medicaid programs that also have similar DRG systems for inpatient hospital 
reimbursement. (27) Government is not the only cost conscious purchaser of 
medical care. Health insurance carriers are adopting alternative payment 
practices and are offering more price-competitive policies in response to 
purchaser demands. One unfortunate result from this price squeeze has been 
the increase in the number of economic transfers of patients from private 
hospitals to public hospitals. Economic transfers - also known as patient 
dumping - have increased in such areas as Boston, Chicago, Washington, D.C. 
and San Francisco. A study of economic transfers in the San Francisco area 
found 63 percent had no medical insurance at the time of transfer, 34 
percent had Medicare or Medicaid coverage and 3 percent had private 
insurance coverage. (28) Exacerbating the problem of economic transfers is 
the fact that the hospitals receiving the transfers are often in worse 
financial shape than the originating hospital. These hospitals usually 
have higher percentages of uncompensated care, higher levels of Medicaid 
and Medicare patients, and lower levels of privately-insured patients that 
can be charged higher rates. 

Uncompensated care charges represent only approximately 5 percent of 
all community hospital charges. The burden of uncompensated care, however, 
is not evenly distributed and falls disproportionately on certain types of 
institutions. Data for the year 1982 show that public hospitals, teaching 
hospitals, urban hospitals and hospitals located in the South provide 
disproportionately high levels of uncompensated care. 

■ Public hospitals account for 20 percent of aggregate total 
charges, and 40 percent of total uncompensated care costs. 

• Major teaching hospitals (public and private) accounted for 
24 percent of aggregate total charges and 36 percent of total 
uncompensated care costs. 

• Urban hospitals (public and private) account for 39 percent of 
aggregate total charges and 49 percent of total uncompensated 
care costs. 

• Southern hospitals represent 31 percent of the Nation's 
aggregate total charges but 48 percent of its uncompensated 
costs. 

• Investor owned hospitals account for 8 percent of charges but 
only 5 percent of total uncompensated care costs. (29) 

Hospitals that combine these characteristics have especially high 
levels of uncompensated care. For example: 



304 



• Major public teaching hospitals account for 5 percent of 
aggregate total charges but 21 percent of uncompensated 
care. 

• Southern large city public hospitals account for 2 percent of 
aggregate total charges and 12 percent of uncompensated 
care. (30) 

In short, public hospitals are bearing a disproportionate share of 
the uncompensated care burden. 

The foregoing discussion has focused on uncompensated care provided 
in hospitals. It is this aspect of uncompensated care for which the most 
extensive data are available. Less is known about the volume and costs of 
uncompensated ambulatory care provided in settings other than hospitals. 
Also, little is known about the extent to which a person's inability to pay 
causes the deferral of care, and the cost and health status consequences of 
such deferred care. 

Several Federal programs, however, address the need to encourage 
ambulatory care services to the medically underserved. The Indian Health 
Service is discussed later in this paper. Other programs include: 

Community Health Centers 

The Community Health Center program supports about 600 grants to 
community health centers in medically underserved areas, about 390 in rural 
areas and 210 in urban areas. While there are approximately twice as many 
rural grantees as urban grantees, funds are split almost evenly between 
rural and urban centers. Community health centers provided primary health 
care services to approximately 5.1 million persons during 1984. Community 
health centers use a sliding scale fee structure and no one is denied care 
because of an inability to pay. Fifty-eight percent of all center users 
had income levels at or below the poverty level and 84 percent were at or 
below 200 percent of the poverty level. (31) 

Migrant Health Centers 

The Migrant Health Program provides comprehensive primary health care 
services to an important subset of the underserved, migrant and seasonal 
farm workers and their families. Access to health care is difficult for 
this group because of its mobility, language and cultural differences and 
low income. Most states consider migrants temporary residents, rendering 
them ineligible for Medicaid. Over 120 grantees are funded, covering 
over 300 rural areas in 35 states and Puerto Rico. They serve over 450,000 
migrant and seasonal farm workers, representing 16 percent of the total 
estimated migrant and seasonal farm worker population in the country. (32) 



305 



National Health Service Corps 

The purpose of the National Health Service Corps is to improve the 
capacity to provide health personnel to the areas with the greatest need 
and demand for health care services and which have been unable to attract 
providers of health care services. To accomplish this the National Health 
Service Corps recruits and places physicians and other professionals in 
health manpower shortage areas. In FY 1984, the National Health Service 
Corps provided primary health care to approximately 2.28 million 
persons — 810,000 people were served by Federally employed physicians and 
1,470,000 by private practice option physicians and private placement 
physicians. A total of 1,291 new, obligated health professionals were 
placed: 139 as National Health Service Corps Federal employees, 448 under 
the private practice option, 495 as private placement physicians and 209 in 
the Indian Health Service. (33) 

Need 

The Task Force has documented the disparities in life expectancy and 
mortality rates between minorities and the White population and has 
identified and studied the leading causes of death which account in 
substantial part for these differences. The need for care, as evaluated by 
a health professional or self-perceived, is one of the best predictors of 
health service utilization. Common measures of health need include 
self-reported health, days of restricted activity per year and bed 
disability days. The Robert Wood Johnson Foundation's survey also 
considered whether a person had had a medical emergency in the past year or 
was part of a family with a seriously ill member. For each of these 
indicators, the Black and Hispanic experience was less desirable than for 
Whites. (Table 7). (34) 

IV. Realized Access 

Health Care Services Use 

The usual unit of measure for utilization of ambulatory care is the 
physician visit. Hospital care use is measured by indicators such as 
admission rates or average length of stay. 

The total U.S. population as reported in the National Health 
Interview Survey visited physicians an average 4.7 times per year 
(Table 8). Annual visits by Whites, Blacks and Hispanics in the aggregate 
were very similar. Variations appear when data are examined by age and 
income. Black and Hispanic children visited a physician less often than 
White children. Physician visits for all groups increased with age, and 
Blacks and Hispanics over 45 visited physicians more often than Whites. 
People with a family income of over $10,000 averaged 5.4 physician visits 



306 





W) 






ai 






o 


CM 




1 — 


CO 




o 


ON 




<: 


r— 


r-^ 


L. 




bJ 


in 


o 


_i 


oo 


00 


CO 


Ld 


CJN 


<c 


CJ 




1— 


CJ 


1 




< 


CD 




Q 


(Js 



c 

CD 
Q. 
CO 



en 






03 



00 



CD 



CM 



CM 
CM 



ON 



Cs) 


LA 


03 


r^ 


NO 


a 


• 


• 


• 


• 


r^ 


<f 


00 


00 


NO 



tH 




O 




tj 




•H 




CD 




Ll. 






* 


4-> 


^ 


u 


4-> 


o 


.— 1 


Q. 


CD 


(U 


(U 


ce 


or 


C4- 


u 


1—1 


o 


0) 


o 


to 


Q. 





* 




t^ 




CD 


1— 1 


(D 


CO 


>- 


o 




•r^ 


4-1 


■a 


CO 


CD 


CD 


2: 


Q- 


ii 


C 


4-> 


■H 


•H 




2 


>^ 







m 


c 


c 


(U 





CJi 


CO 


(-1 


^ 


(U 


(U 


E 



SI 




4-> 




•H 




s 




(H 




QJ 




J3 




e 







* 


s 


CO 




CO 


jr 


(D 


4-> 


c 


•H 


1— 1 


s 


1—1 




t— 1 


CO 




(U 


CO 


•H 


D 


^H 





■H 


•H 


E 


(-1 


CD 


(D 



Q_ UJ 



s? 



>s 




-u 




■H 




> 


* 


•H 


* 


-l-> 


^— ^ 


CJ 





cC 


00 




C3N 


•0 


r— 


(U 


1 


4-1 


00 





r- 


•H 


CT\ 


(H 


^— 


-(-> 


^^ 


CO 




(U 


U 


q: 


CD 




(D 


it- 


>- 









U 


CO 


<u 


>^Q_ 


CD 




Q 





Ch 




(U 




Q_ 






* 


CO 


* 


>^ 


^-^ 


CD 





Q 


03 







>> 


^ — 


4-) 


1 


•H 


CO 


1— 1 


r^ 


•H 


ON 


XI 


^ — 


CD 


^.^ 


CO 




•H 


u 


Q 


CO 




<D 


■D 


>- 


CD 




CD 





CD 




(D 




X 









4-) 


E 




CO 


CO 


c 


CO 


CO 








u 


•H 





(4 


<i 


CD 




E 


c 


< 









(D 


4-1 


4-! 


t4 


■H 





JZ 


a 


s 


CD 




q: 


■a 




c 


■0 


CD 


(D 




4-> 


^ 


CD 





■D 


CD 


a 


^ 



(4 
o 
a. 



CD 

•H 

o 

(D 
Q. 

tn 

c 
o 



CD 
■D 

c 
o 



c 
o 
en 

c 

SI 

o 

r-) 

■o 

O 
O 



4-1 


CD 


C 


(4 


— 1 


•H 


ID 


Q. 


> 


JD 





QJ 





(D 


t4 


OC 


Q. 


1— 


* 




* 



C_) 

cc: 

O 



CO 



c 

CD 
CL 
CO 



u 
o 
tl- 

co 

C4 

o 

-u 

CD 

o 

•H 
T3 

c 



4-> 



CD 
(D 



307 



per year, while those with family incomes of less than $10,000 averaged 4.6 
visits. This difference of approximately one visit between income groups 
continued across racial groups, with Whites averaging more visits at both 
income levels. (35) 

The masking effect of aggregate statistics is evident in the 
physician visit indicator for Hispanics. While Hispanics averaged 4.4 
physician visits per year, Cubans averaged 6.2 visits per year and Puerto 
Ricans 6.0 visits, well in excess of the White average of 4.8 visits, while 
Mexicans averaged 3.7 visits per year. (36) 

The site of physician visit varies between Whites and Blacks. Whites 
were more likely to see a doctor in his office (69.2 percent) than Blacks 
(58 percent). Over one-fourth of Black physician visits occurred in 
hospital clinics or emergency rooms, but only 11.2 percent of White 
physician visits took place there. Whites are more than twice as likely to 
consult their doctors by telephone than Blacks. The higher one's income, 
the more likely one is to have an office visit; the lower one's income the 
greater the likelihood that a physician visit will take place in a hospital 
clinic or emergency room. (Table 9) (37) 

There is a pattern of lower use of preventive health services among 
minorities. Prenatal care provides an opportunity to identify and treat 
medical problems and to educate patients about the effects of diet, 
smoking, alcohol and drug use on the fetus. Close to 80 percent of White 
women receive prenatal care during the first trimester of pregnancy, while 
only 62 percent of Black women receive early prenatal care. In the 22 
states which report Hispanic birth information, only 60 percent of Hispanic 
mothers receive first trimester prenatal care. (38) 

The lower number of minority physician visits for children is 
reflected in lower vaccination rates. Sixty-seven percent of White 
children under four had been vaccinated against measles in 1983 versus 57 
percent for minority children. Even larger differences in vaccination 
status were observed for rubella, DPT, polio and mumps. (39) 

Dental services are used less frequently by minority populations than 
by Whites. Approximately 56 percent of all Whites four years of age and 
over visited a dentist in the previous year during the time period 
1978-1980 compared to 37 percent for Blacks and 40 percent for 
Hispanics. (40) 

During this same period, hospitalization rates for these population 
groups did not vary substantially among groups. When adjustment is made 
for age, 10.3 percent of Whites, 11.1 percent of Blacks and 10.2 percent of 
Hispanics were hospitalized at least once in the previous year. Within the 
Hispanic population Mexican Americans were hospitalized least (9.6 percent) 



308 



00 



CD 



UJ 



o 



< 




L. 




Q 




Z 




«C 




»t 




X 




UJ 

to 


o 


LiJ 


CO 
ON 


U 


'^ 


< 


^ 


<: 


ON 
ON 


•« 




Q 
a 


CD 

ON 


CD 


m 

CD 
CO 






CJ 


■a 



< 

Q- 



>- ^ 

a: § 
t^ Q 



CD 

I — i 

> 

<: 
I— I 

I — I 
tn 

>- 



o 
a: 

LjJ 
CO 



CJ 
CC 

i— I 
CO 



to 




CJ 




1— 1 




_J 2 


<t 


_i <: 


• 


<: Q. 


<J- 


to 




1— 1 




X 









LA CM 00 CSI 
• • • • 

^N <t LTi CO 



NO r^ 

• • 
r^ lA 



-H CO 

.— I <D 

<C O 

C0 



NO 



GO 



cNj 03 cjN r^ 
• • • • 

rA <}• lA NO 



lA lA '— r«^ 

• • • • 
<t <t lA NO 



Q. 

o 

Q. 



to 

I — I 

a: 



CJ 

<1 

CC 

<: 



rA LA CNj r<^ 

• • • • 

<j- <r LA NO 



00 CM 

• • 
PA lA 



o 00 

lA ^ 



LA <»• 



»- LA 

• • 

<t LA 



CM lA CM <t 

• • • • 

<t <!■ LA NO 



CO 

CO r— 

(D 

>- t-i -^ o- 
D O- NO 

C TD I I + 
•H C r-- lA LA 

=) T- <* NO 

en 
< 



<J- LA 



NO NO 

• • 

lA vf 



lA NO 

• • 
LA 0- 



CD <3- 

<f LA 







<D 






. — 1 




CD 


CO 




. — 1 


E 




CO 


CD 


X 


s: 


L- 


(U 






to 







LA <t 



O 

O 
(D O 
E - 
O O 
O <^ + 
C *«• O 

>^ CD - 
■— I TD O 
•H C r- 

E ^ ■«* 

CO 



0) 

> 
to 



CD 
0) 

3: 



CD 

c 
o 



(U 




^ 




*J 




t4 




o 




u- 




co 




■p 




CD 




Q 




CO 


• 


c 


^— s 


CD 


NO 


CJ 


r^ 


•H 


lA 


t< 




CU 


1 


E 


<r 


<C 


CO 


0) 


^^ 


-P 


to 


•H 


X 


XZ 


Q- 



•a 





o 




CD C 




rH O 




QQ -H 




4-! 




- CO 




o o 




•H -H 




C r-l 




CO X) 




Ql D 




CO Q_ 




•H 


• 


X X 


(D 


X 


E 


tJ Q 


O 


O 


O 


ij- ^ 


c 


CD 


•H 


CO <}• 




tj T— 


>, 


o 


1—1 


■u • 


•H 


CD O 


E 


O Z 


CD 


•H 


C4_ 


"D - 




C O 


C 


•H T- 


s 




o 


jz CO 


c 


-U 0) 


J^ 


.—1 -l-l 


c 


CO (h 


3 


(D (D 




X to 


CO 




(D 




X) 


• • 


:3 


UJ 


i-H 


CJ 


o 


on 


c 


=) 


1— 1 


o 


* 


to 



309 






OS 



< 



Q 




Z 




<C 




Ld 




CJ 




< 




cc 




^ 




X 




UJ 




W) 








LiJ 


o 

00 
On 


Q 


CO 
(D 

-4-1 




CO 






1 — 
1 — 1 


IT) 


1 — 1 

> 


■a 

CD 


o 


•H 

c 


UJ 




CJ 

—1 

Q- 


• • 

CO 

CD 


:^ 


■D 


OD 


"O 


on 

1— 
1 — 1 


O 

JZ 
<D 
CO 

3 



CJ 5 



Z3 

m 
I — I 

I— 

Q 



CJ 



LO 



=a 



c 
s 
t-i o 
CD C 

c 

O ID 



>>4-l 

-(-> C 

CO ZD 

c "a ^ 

CD C 4.) 

Q. 1— I •— I 

E CD 

O U (U 

CJ o x: 







c 


UJ 


JD 


o 


CJ 




•H 


< 


CD 


4-> 


—1 


C 


CO 


C3. 


o 


-(-) 




JZ 


r— 1 




Ci 


D 




CD 


CO 






c 




QJ 


o 




1 — 


CJ 



C-l 

o 



£3. C 



E 
O 

_ o 
:r CJ UJ cc 



c 




CO 


CD 


•iH 


O 


o 


■H 


•■H 


U- 


CO 


<*- 


>nO 


J= 




Q_ 





CJ 

I— I 

I— 

CO 

I— I 



CJ 



CJ 



CJ\ 



T— LA <J" r^ 

• • • • 
^ so u-\ 1-^ 



CM LA M3 (33 

• • • • 

o o o ^ 



r- T- C^ T- 
• • • • 

CD ■r- O O 



t— Cs) CM CJN 

• • • • 
CD r- O CD 



CD r^ rA (Nl 

• ■ • • 

CM <3- CM O 



r^ <j- r-~ r^ 



C>4 VO O LA 

SO MD r-- r^ 



CO 

CD 
CD 

>- 



CD 

cn 
< 



C-i 
(D 
TD 

c 

Z3 



<t so 

I I + 

LA lA LA 

T- <)• SO 



CO r^ 

LA LA 



r^ so 

C3 O 



SO r-- 

o o 



o o 



CM -^ 

• • 
LA ^ 



CM O 



SO OS 
so SO 



X 

CD 

to 



CD CO 
— ( E 
CO CD 



T- CM 

• • 
LA O 






so 1^ 

■ • 

o o 



r- Csl 

• • 
lA LA 



CM SO 



^ LA 
^ CM 



CM 



c^ 00 

so LA 



CD 

o 
CO 



(D Ji 

-1-1 o 

■H CO 

JZ ^ 

2 CD 



lA O CM 

« • « 
00 lA <t 



r~- rA ^— 

• • • 

CD a o 



<t 00 CO 

• • • 

o o o 



CTs 00 CM 

* • • 

OS CM -a- 



<)• ^ rA 

• • « 

r^ CM OS 



1^ CD LA 



rA OS O 

SO so r-~ 



■fcfl- 



t-l 

>^ <D 

^ TD 
•H C 

E =1 

CD 



OS 
CJs 

OS (D 
^ (-1 
<t O 
CM E 
•fcft- 

u 
I o 

o o 

D O 
O O 

CD LA 
t- CM 



CO 

(D 

CO • 
o ^ 

Q. O 
(-1 CD 
D —I 
CLJD 

^ U 
CO o 

o 

TD -P 
CD -H 
E ^ 
C S 
O 

C C 
CD 
u ^ 
CD 4-> 
SZ 

-U t-l 
O (D 
JI 
TD -U 
C O 
CO 

03 
CO CD 
O 
CO 
t-l 

XI 

c 

CO 

CD 

E 
O 

o 

c 



c 

<D 

E 
-P 
C 
•H 

o 

Q. 
(D. 
CD 

U 
O 



CL >^ 

D CO ^ 

O --H -H 

ti --I E 

CD CD CO 

O it- 

TD 

•H CD C 

CD "D S 

Q. 3 O 

0) ^ C 

Q- C C 

•H 3 
CJi 

C -tJ CO 

■H O CD 

C TD 
D 

CO r-H 

CD O 
_ O C 
I— ( Q I— I 
CO JD O 



TD 

D 






(D 

> 
(4 
D 
CO 



CD 
0] 



CO 

c 
o 

•H 

4-> 
CO 



CD 



o 

Ct- 

co 

CD 
Q 



O 
00 
OS 



CO 
<D 

4J 

CO 

4J 

cn 

TD 

CD 



cn 

> 

CO 
CD 



ID 
O 

c 

■H 
CO 



CO 

> 
u 

(D 

c 



=a 



CD 




E 




D 




i-( 




O 


1 


> 


-d- 




<t 


M 


t— 


4J 




■H 


• 


CO 





•H 


z 


> 




c 


a 


CD 


T- 


•H 




o 


CD 


•H 


CU 


CO 


•H 


>^ 


P 


JI 


(U 


Q- 


C/) 


• • 




U 




u 




q: 




rD 




a 




(/) 





310 



O 



a 

00 



OS 






CQ 



05 

CO 
•□ 

(D 



5 =) 



C_) 


CD 

-a 


Q_ 


■o 


_ 


rH 


Q 

bJ 


O 


1— 
ZD 

n 

Q 


to 
o 


<: 

I 


CT) 


1 

LlJ 


1 — 1 

X 



c 

CD 

m 

•H 

(D 



O 

Q. 





C 


r-j 




CO 


• 


^ 


u 


CSJ 


CO 


•H 


^- 


-Q 


C-i 




13 


CD 




(LJ 


E 





o 

-u c 

Ci CO 

(D a 



X 

CD 
S < 



o 

•H 

c 

CD 
Q. 
CO 

•H 

I 

c 
o 



t>^ (J 
< CO 

CL 

CO 



OD 



CSI 

CO 

.— I 0) 

^ o 
<: CO 



3 
o 

Q- 



C_) 



VO 
OS 



CSJ 

O 












LO 


CO 


1— 1 


c 


ce 


o 


LU 


CO 


1— 


(h 


OJ 


CD 


< 


(D. 


EC 




<c 


1— 1 



V- CSJ 



O OS 
00 r^ 



C3S tA 

• • 

SO CN 



CO -a- 

• • 

r-- CSJ 



OS 00 

• • 

00 CsJ 



OD CO 

• • 

CO ^ 



r-- OS 

• • 

CO T- 



r-^ C3S 

00 ^ 







CD 






.— 1 




OJ 


CD 




1 — 1 


g 




CD 


QJ 


X 


s: 


Li. 


(D 






in 







LA so 
rA CN 



CM 00 

• • 

lA 00 



rA r^ 

^ CO 



LA (Nl 

• • 

CSJ CTn 



OS so 

• • 

CM CJs 



^ CO 

• • 

cvi OS 



<t CO 

• • 

CSI 0\ 



<t CD 

• • 

<M C3S 



O 

O 

o 



ID 

u 
o 



o 



ID 
E 
O 
O 

C -tfi- o 
h-i o 

>^ (D »■ 
^ -D O 
•H C r- 

E =D *«• 

CO 



<t CJs lA 

• • • 

<NI CN ^ 



^ r- r^ 
• • • 

tA lA CO 



T- O O 

• • • 

OS T- OS 



O SO (Js 

• • • 

O r- OS 



rA fA rA 

• ■ • 
r- T- O 



SO r^ CO 

• • • 

r- T- c^ 



LA r-^ 00 

• • • 

^ T- Ov 



CSI so CO 

• • • 

r- T- c^ 



CD 
CD 



^ 









•H 






C-I 


E 






O 


CD 


CO 




E 


b. 


t-i 








CD 




(-1 


li_ 


CD 


CO 


O 


O 


>N 


d 








CO 


CO 


C 


C3S 


(D 


(-1 


o 




>^ 


CO 


•H 


C-i 




0) 


-t-> 


CD 


r— 


>^ 


CO 


■D 


T— 




o 


C 


1 


CsJ 


3 


::3 


OS 


T— 


TJ 









CSI 

CSI 



CTs OS 



tA fA 



rA LA 

• • 

00 OS 



LA O 



00 o 

CSI 



so <t 



CO rA 

fsl 



so LA 



00 CSJ 

CSI 



00 CSJ 
CSI 



CO 

D 

-u 

CD 

4-1 XJ 

m o 
o 

^ en 
-u 

^ t-i 

CD o 
(D 

c 

TJ CD 

(D I— < 

> ^ 



C-I 

o 
o 
a. 

u 
o 



■H 
CD 

o 
tl 

(D 
Q- 



c 
o 



3 

CL 

o 
a. 

-o 

CD 
N 



CO 

c 
o 



3 
-u 

■H 

4-> 

CO 

c 

•H 

c 
o 

c 

c 

CD 



> 

■H 

o 

—I 

CO 
-u) 

o 



(D 



c 
o 



3 

JD 
•H 

u 

4-1 
CO 
■H 

■o 

(D 

CO 

CD 



en 

•H 

kl 
o 



o 
o 

(D 

E 

4-> 

o 

CD 

C-I 



CJs 
CJs 



c 
CO 

CL 

CO 



CO 
(D 

4-) 

CO 

>^4-> 

J3 C/5 



>- c 

2 

>s O 

c 
c 



en 

■H 

c-< 
o 

CJ 

•H 

c 
CO 

Q. 

CO 



<D 

JO 

4-> 



■a 

(D 



c 
CO 

CO 

(D 
CJ 
CD 

u 

C4 

(D 



ID 



CO -H 
D C 

CD (D 
(D 4J 

<: 



CO 

(D 
1-H 

o 

c 



o 

CD 
CL 

CO 

c 
s 
o 
c 
^ 
c 

CO 

(D 
T3 

D 
I— I 
o 

c 



Csi rA 



311 






a 
CO 

ON 



iS) 



OS 



Si ^oT 
o , r-- 

X T^ o 





si 


(U 

4-) 




CD 


^ 


Ql^ 


4-1 


CQ 

<: 


IBER OF 
OR MOR 


■H 

c 



^ " 



< 


s 




> 


to 
o 

on 

UJ 
Q. 




<: 


r— 1 

o 







1 — 


a: 
o 


O 




^ 


Q 




CO 


1 




1 — 1 

X 


Ld 




z 


U 






<: 







c 

03 

a 

05 



o 

Q. 

U1 





c 


O 




CO 


« 


c 


o 


CTs 


CO 


(-1 




13 


(D 




CJ 


E 





o 

4J C 

tj CD 

OJ u 

■J -H 



s: < 



— H -H 

rH C 

<: CO 

Q. 
CO 



c 

CD 
Q. 
03 



O 
CD 

m 



CSJ 

03 
-—I <D 
—I O 
<C CD 



c 
o 

•H 

^ -(-) 

CD CO 

4-> I— I 

o n 

I— Q. 

o 



CJ 



CJs 



00 



00 

CD 



CD 



00 



CO 



Ul 


03 


1 — 1 


c 


q: 


o 


UJ 


03 


1 — 


U 


OJ 


(D 


<c 


a. 


cc 




<: 


•—{ 



t^ \D 



vo r-~ 

• • 



• • 

o 00 



t^ CT\ 

• • 

r^ ON 






cjN 00 



C3N CO 



.—1 


E 


CD 


(D 


X 2: 


Ll. 


(D 




tn 





• « 



lA 



o\ 



lA CJ\ 

• • 

ON NO 






00 NO 

• • 

^ C3N 



ON 00 

• • 



t^ CD 

• • 

o CO 



r^ lA 

« • 



o 
a 
o 



u 
o 



CD 
E 
O 
O 
C -fefl- □ 

l-H □ 

>, OJ - 

rH "O O 

•H C: r- 

E =3 -e^ 
CO 



<!■ ON NO 



NO NO LA 

• • • 

r^ CJN NO 



r- r^ r- 
• • • 

CJN CJN NO 



lA 00 CNJ 

• « • 

o ON r-~ 



t^ rA KN 

• • • 

r- T- a 



<J- CO ^ 

• • • 

On CO CO 



CD rA csj 

• • • 

o o CO 



T- rA csj 

• • • 
O ON CO 



CO 

0) 



r- 1 






03 


•H 






U 


E 






O 


CO 


03 




E 


Ll. 


t-i 








CD 




Ch 


U- 


03 


03 


O 


o 


>, 


U 








CD 


03 


c 


ON 


OJ 


U 


o 




>< 


CO 


•H 


u 




03 


+J 


(U 


r- 


>^ 


CD 


"O 


^ — 




O 


c 


1 


CM 


■D 


ZD 


On 


T— 


■D 









^A o 

• • 

1^ <t 



LA LA 

• ■ 

00 <J- 



O 00 

• • 

NO CM 



ON fA 

NO rA 



O CO 

• • 

CJN O- 



CO rA 

■ • 

NO rA 



• • 
r^ rA 



o 1^ 

• • 
r^ rA 



03 

-p 

CO 

4-> TD 
LO O 

o 
en 



4.) 
r— I 
CO 

0) 

X 



c 

03 



U 
O 
O 
Q. 



"O 

03 --H O 

> —I 

•rH 03 fcj 

O -H 

O X CO 

Ch LJ Li- 
03 



C 

o 

■H 
-P 

CO 

f-H 
13 
CL 
O 
Q. 

■a 

03 
N 



CO 

c 
o 

■H 
4-> 

4.) 



03 

C 
•H 

c 
o 

c 

c 
CO 



CD 
4-> 

o 

4-) 

03 

x: 



c 
o 

•H 
4-> 
D 
jD 



03 
•H 

■a 

03 
Di 
CD 

03 



0\ 

c^ 



en 
■i-( 
u 
o 

o 
—I 

c 

CD • 
Q. 
03 • 



•H 



"D - 

O r- 

JZ 

4-) >^ 

03 .—I 

E 3 
O 

4-) 

O 14- 

03 O 

U 

•H 03 

■O CD 



c 

CO 
Q. 
03 

•H 
X 



c 

o 

c 

.NC 

c 

3 T3 
03 
-O -H 
C l^ 
CD -H 
O 



03 

4-1 



03 
03 

4J 

CD 

>^4-l 

J3 1/3 

T3 -O 

03 03 

4-1 4-1 

03 -H 

n c 

■D 
CO 03 

SI 

4-1 
CJi 



0) 

a 

03 

c 

2 
O 

c 
.^ 

c 

0) 
03 



-a n 



CM lA 



312 



and Cuban Americans the most (12.2 percent). The age-adjusted average 
number of days in the hospital for those with one or more hospital stays 
was lowest for Whites (8.4 days), followed by Hispanics (8.8 days) and 
Blacks (11.0 days). The number of hospital days experienced is inversely 
proportional to family income, family education and perceived health 
status (Table 10, Table 11). (41) 

If the health status of minority groups were similar and access to 
health services were equal, one would expect utilization of different types 
of health care services to also be similar. The Task Force, however, has 
demonstrated that there are significant differences in life expectancy, 
morbidity and mortality among major population groups. One would expect, 
based on health status alone, higher utilization rates for minority 
population groups than for Whites. Thus, the differences in the use of 
health services presented in this section are magnified in importance. 

Utilization measures indicate whether people receive services but do 
not relate how satisfied patients are with their care. The 1982 Robert 
Wood Johnson survey, summarized in Table 12, finds Black and Hispanics to 
be less satisfied with their medical care than Whites for most aspects of 
recent medical visits. They reported lower levels of satisfaction for 
travel time, waiting time in the office, time with the doctor, the 
information they received, the quality of care received and the overall 
quality of the visit. 

V. Access to Health Care: American Indians and Alaskan Natives 

Definitional questions and data availability complicate consideration 
of access-related indicators for the American Indian-Alaskan Native 
population. 

The 1980 Census identifies 1,534,000 Americans as American Indians 
and Alaskan natives. A culturally and historically diverse group, the 
Native American population shares some of the demographic characteristics 
of other minority groups. Census data show it to be a young population, 
with 10.1 percent of the population less than five years old while only 5.2 
percent, less than half the national rate, is aged 65 or more. (42) 

For those in the civilian labor force, unemployment rates are high, 
13.2 percent in 1980, while median family income is low, 84 percent of the 
national 1979 average. Twenty-seven and one half percent of all Native 
Americans are below the poverty line. (Tables lA and 2A) . 

American Indian population is concentrated in the 32 Reservation 
States. A State is considered a Reservation State if the Indian Health 
Service (IHS) has responsibilities within the State. There are currently 
32 Reservation States. Maine, Pennsylvania and New York were added as 
Reservation States in 1979; Connecticut, Rhode Island and Texas in 1983; 
and Alabama in 1984. The 32 Reservation States are: 



313 



on 
or 
o 

I— 
<: 






u. 
to 



< 

to ^ 

< 

CO 



Q 

:^ 

< 

CD 



CD 

a 
en 



o 

CD 

cn 



CTs 
CM 






c?\ 



CM 






CM 



00 
CM 



CM 
CM 






CM 



<3- 
CM 






CM 



CO 

CM 



c^ 



r^ 



CD 
CM 






CO 

o 

•H 

c 

CD 

Q. 
CO 



cn 
c 
o 



CD 
U 

CO 

CJ 

CO 



o 

4-> 

CO 
CO 

(U 

o 
o 

< 

CJl 

c 

■l-> 
CO 

■D 
Q. 



CD 

C 

CO 



o 

CO 



CJ 





4-> 
















•< 


CO 


•H 
















• 


4-1 


CO 
















z 


>^ O 


•H 


















r-H CD 


> 
















• 


(D CL 










T) 








ce 


4-) CO 


-— 1 




Q) 




(U 


CO 








(D < 


CD 




E 




> 


4-> 






•< 


.— 1 


O 




•H 




■H 


CO 






c 


Cir: 


■H 




1— 


k< 


0) 


o 









S 4-> 


■a 






O 


o 


(_) 






CO 


O -H 


CD 




cn 


4-) 


OJ 








Ch 


CJ s 


s: 




c 


o 


en 


4-) 




4J 


(D 








•H 


o 




(D 




■H 


■o 


JJ TJ 


4J 


QJ 


4-> 


o 


c 


J^ 




m 


c 


O (D 


c 


E 


■H 




o 


O 




•H 


<t 


Z -H 


<D 


•H 


CO 


^ 


•H 


O 




> 




C(- 


O 


1 — 


s 


4-) 


-U 


£3- 








4-1 CO 


(U 






•^ 


CD 


1 


>^ 


r— I 


• • 


C -H 


a: 


1— 1 


(D 


3 


E 


U- 


4-) 


.—1 


bJ 


CD -(-> 




(U 


O 




t4 


o 


•H 


CO 


C_) 


O CD 


C4- 


> 


•H 


0) 


O 


1 


^H 


u 


q; 


t-i CO 


o 


CD 


It- 


E 


14- 


4-> 


CO 


CD 


^ 


(D 




t-i 


Ct- 


•H 


c 


D 


3 


> 


CD 


Q. 




h- 


o 


1 — 


HH 


O 


C3r 


o 


CO 



314 



Alabama 

Alaska 

Arizona 

California 

Colorado 

Montana 
Nebraska 
Nevada 
New Mexico 
New York 
North Carolina 



Connecticut 

Florida 

Idaho 

Iowa 

Kansas 

North Dakota 
Oklahoma 
Oregon 

Pennsylvania 
Rhode Island 
South Dakota 



Louisiana 

Maine 

Michigan 

Minnesota 

Mississippi 

Texas 

Utah 

Washington 

Wisconsin 

Wyoming 



Of the total American Indian population, 1,295,000 reside in these 
states. (43) 

Unlike other minority populations, the Native American population is 
more rural than the general population. Its physical isolation contributes 
to the types of health care problems which characterize it. 



The Federal responsibility fo 
makes this population unique among 
when one considers its access to he 
care delivery arises from the histo 
Congress and individual tribes. Th 
legislative basis for IHS services, 
transferred from the Department of 
Service. The mandate of the Indian 
comprehensive health services to Am 
living on or near Federal Indian re 
country such as Oklahoma and Alaska 



r providing health se 
minorities and raises 
alth care. This Fede 
rical treaty relation 
e Snyder Act of 1921 

In 1955 Indian Heal 

the Interior to the P 

Health Service is to 

erican Indian people 

servations or in trad 



rvices to Indians 
unique questions 

ral role in health 

ships between 

forms the 

th programs were 

ublic Health 
provide 

and Alaskan Natives 

itional Indian 



The IHS estimated its 1985 service population as 962,000. Thus, not 
all Indians in Reservation States fall within the IHS service population. 
Approximately 90 percent of the IHS service population lives in 11 states: 
Alaska, Arizona, Minnesota, Montana, New Mexico, North Dakota, Oklahoma, 
South Dakota, Utah, Washington and Wisconsin. In these states, 92 percent 
of Indians are served by the IHS. (44) 

The substantial number of self-identified Native Americans not 
included in this total either live outside the geographic range of its 
service units or are not members of Federally recognized tribes entitled to 
its services. 

There are two principal sources of data on health status and 
utilization measures for Native Americans. The Health Interview Survey 
(HIS), the source of already cited information for Black and Hispanic need 
and utilization indicators, provides information on need-related access 



315 



indicators and utilization measures. HIS data are drawn from household 
interviews in about 42,000 households representative of the civilian 
non-institutionalized population. The small percentage of Native Americans 
in the general population results in a small sample in the HIS, a sample 
which despite the limitations inherent to its size represents the Native 
American population as a whole and not just that portion served by the IHS. 
Data from the 1978 Health Interview Survey show Native Americans reporting 
levels of health status and utilization which diverge from the majority 
patterns. (Table 13) (45) 

The only source for data on Native American coverage by Medicaid and 
Medicare is the HIS. In 1978, the HIS reported that 8.3?o of all Native 
Americans were covered by Medicare and 17.4?o were "probable Medicaid" 
program participants. The extent to which private insurance, Medicaid and 
Medicare pay for Native American health care is not fully known. 
Similarly, the degree to which Native Americans living within IHS service 
areas seek and receive care from sources other than the IHS and how such 
care is financed is not now known. (46) 

Within its service areas the IHS directly operates 47 hospitals, 80 
health centers and more than 500 health stations and satellite clinics. 
Four hospitals and 292 health clinics are operated under the tribal health 
delivery system which is administered by tribes and tribal groups through 
contracts with the IHS. Supplemental services not available through IHS' 
direct or tribal facilities are purchased from appropriate providers under 
contract. (47) 

During FY 1984, the IHS directly, through the tribal health program 
and through contracted purchase of services, was responsible for 4,232,000 
outpatient visits and 103,000 hospital admissions. (48) The number of 
outpatient visits has remained relatively steady since 1981.(49) The 
number of hospital admissions peaked in 1978 and has varied within a narrow 
range since then. The average daily hospital patient load in IHS, tribal 
and contract hospitals continues to decrease in large part because of a 
drop in the average length of stay which was 9.3 days in 1970 but stood at 
4.9 days in 1984. (50) 

Over the history of the IHS there have been significant improvements 
in the health status of the American Indian people. The infant mortality 
rate has dropped to 13.8 per thousand live births compared to the U.S. 
overall rate of 12.8. Maternal mortality declined to a level at or below 
that of the U.S. as a whole. Infectious diseases such as tuberculosis have 
been brought under control. As a result of addressing the acute problems 
which faced this population, the health problems of the American Indians 
have changed over the past 30 years. Chronic diseases associated with 
aging have become an increasingly common cause of death. Accidents and 
injuries is this population's second leading cause of death. Alcoholism is 



316 





c 








CO 








a 








■H 








t4 








(D 


VO 


CM 




E 


• 


• 




< 


<* 


00 






CSI 


^ 


o 

1 — 

•a: 


> 

•H 
4-1 

CD 

Z 







CM 



o 



CM 



On 



t^ 



CD 






=5 



CJ 

I— I 
q:: 
bJ 

s: 
<: 

UJ 

> 
I— I 

I— 
< 



•H 

JZ 



c 

CD 
£3. 
CO 
•H 

X 

I 

c 
o 



CD 






r^ 



O 



LP, 



CM 
CM 



a: 
o 






u 

CD 
(D 

>- 

U 

cu 
a. 

CD 

>^ 

CO 
Q 

>. 

■H 
> 

•H 

o 

•St 

■a 

CD 



C-l 

en 



-u 

■H 

> 

a 



c 
o 

C_) 

4-> 
•H C 

s o 

■H 
CO 4-> 

c CO 

O +J 
CO -H 

^1 E 
0) -H 

Q I 



U 
O 



u 

■H 
CD 



0) 



TJ 




0) 




> 




•H 




0) 




o 




u 


JT 


(D 


■I-) 


Q. 


1— 1 




CD 


CO 


(U 


c 


X 


o 




CO 


tj 


(-1 


o 


(D 


o 


Q_ 


Q. 



-P 




•H 




CO 




•H 




> 




c 




CO 




•H 




o 




■H 




CO 




>^ 




^ 




Q- 




CO 






M 


JZ 


CO 


-l-> 


<D 


•H 


>- 


2 






-u 


CO 


CO 


c 


CD 


o 


Q_ 


CO 




C-i 


OJ 


CD 


^ 


Q. 


-p 



c 




CO 




•H 




a 




•H 




CO 




>^ 




^ 




Q. 




CD 




(4 




O 




s: 


tJ 




CD 


(h 


CD 


o 


>- 


ITi 


+J 




CO 


x: 


CO 


4-> 


O- 


■H 




s 


c 




■H 


CO 




c 


CO 


o 


-I-) 


CO 


•H 


(-1 


CO 


Oi 


■H 


0_ 


> 



t«^ 




lA 




CO 




4J 




■H 




CO 




•H 




> 




1— 1 




CO 




4-1 




c 




(D 




Q 




tu 




C-l 




o 




2: 




(4 









CD 




C 









jr 


(4 


4-) 


CD 


•H 


(U 


2 


>- 


CO 


4-> 


c 


CO 





CO 


CO 


a. 


!-• 




(D 


c 


0- 


•H 



lA 



c 
o 

CO 

u 

0) 
CL 

(4 

(D 



CO 



CO 

4-> 

c 

(D 
Q 

C4 

li- CD 
O d) 

>- 
(4 

CD tj 
XI !D 

e Q. 

3 



>s 




CD 




4-1 




1 




1 

4-) 




(4 









X 




LO 




0) 




t4 









2: 




(4 











CO 


CD 


Q) 


C 


■0 










CO 


X 


■r4 


4J 


Q. 


•H 


LJ 


s 






1 — 1 


CO 


CD 


c 


4-> 





•H 


CO 


Q. 


^ 


CO 


(D 





Q- 


X 



o 

CO 



317 



an underlying cause for many accidents and this disease, by itself, is a 
major contributor to death among American Indians. Many of the deaths 
associated with these causes are preventable. A central focus of the IHS 
has always been health promotion and disease prevention. Efforts targeted 
at reducing injuries and health risk factors are underway throughout the 
program. (51) 

VI. Asian-American Access to Health Care 

The 1980 Census identified 3.7 million Asian-American/Pacific 
Islanders in the American population. Like other minority groups, this 
population has a young age-structure, with a greater percentage of the 
population under the age of 5 and a smaller percentage over the age of 65 
than the general population. In the aggregate, it differs from other 
groups in several factors relating to potential access to health care. Of 
its members aged 25 years or more, 32.9 percent have completed four or more 
years of college, compared to a national average of 16.2 percent. There is 
less unemployment among this group. In 1980, of persons in the civilian 
labor force, only 4.7 percent of asians were unemployed. The 1979 median 
family income of $26, 456 was 33 percent above the national average. (52) 
Because Asians make up less than two percent of the nation's population, 
recent systematic information on their insurance status and use of health 
services is not readily available. 

The term Asian American comprises a number of diverse groups, the 
largest of which include: Japanese, Chinese and Filipinos. Although the 
Asian population has become more geographically dispersed, it remains 
highly concentrated in the West, with 56 percent of the total. Asians as a 
whole lead an urban existence. Ninety-seven percent of the Chinese live in 
urban areas, followed by Filipinos and Japanese, both 92 percent compared 
to 71 percent for the White majority . (53) 

As Table 14 illustrates, there is considerable variation among Asian 
American subgroups in terms of percentage native born, income and 
percentage of families below the poverty level. (54) 

Within the Asian American population is a subgroup which has been 
recognized as having special problems in seeking and gaining access to 
health care. Between 1975 and 1982 roughly 1.4 million Indochinese 
refugees migrated from Cambodia, Laos and Vietnam, with 580,000 ultimately 
settling in the United States. (55) Since then, the annual number of new 
Southeast Asian immigrants has dropped and stabilized: in Fiscal Year 
1983, there were 61,000 Southeast Asian immigrants, which represented 65 
percent of total refugee immigration. (56) 

These immigrants for the most part came to the United States with 
little preparation, few resources and no realistic hope for a return to 
their former homes. The stress of their recent lives is compounded by 
adjustment to new lives in very different environments. Language, 



318 



CQ 
< 



q: 

UJ 
Q 

z 
<: 

_j 



o 

<: 



ID 

<: 






I — I 



< 

or 






o 
t/1 



o 





c 




CD 




O 




E 




03 




Ln 


1 

03 




e 


c 


03 


CD 


3 


•rH 


CJ 


C 



03 

s c 

03 CO 



CD 

4-> 03 

03 e 

■H CO 

> c 



c 

C 03 
03 -H 
■H T) 

m c 
<: i-H 



c 

03 
03 
U 

o 



o 

c 






o 

<t 
I— 
<: 

Q 

a 

00 



03 
03 
03 

c 

03 
Q. 
CO 

n 



J3 
CD 




•rH 


ON 


t-i 


P- 


03 


0\ 


> 





CJN 



MD 







I^ 


o? 


t^ 


• 


^ 


lA 


csi 


VO 


<t 


t— 



OD 


<>J 




^O 


<!• 


lA 


ON 


cs) 


• 


•s 


*s 


r-~ 


VO 


<r 


CN 



• 


s? 


CO 


CNI 


o 


C3 


<NI 


o 


<r 



CSJ 



CO 






Csl 



CN 



\o 


ON 


T— 


lA 


• 


ON 


<t- 






(=) 




OvJ 



NO 


lA 


ON 


CJN 


• 


<t 



• 


S.? 


NO 


o 


O 


rA 


r^ 


f^ 


CSJ 



NO 



NO 
CM 



ON 
CM 
CM 



o 
rA 



NO 


o 


CO 


r^ 


• 


NO 


t^ 


•> 




<J- 




CM 



NO 



• 


o? 


CO 


CO 


LA 


r~~ 


CM 


1^ 


CM 



c> 



CM 



NO r- 

CM CM 



03 
OnJ 

00 



NO 
ON 









LA 


^— . 


□ 


^ — 






<r 


OD 


r~ 


<r 


• 


a? 


LA 


• 


• 


O-J 


00 


LA 


o 


C3 


CSI 


o- 




»^ 


rA 




KN 


s — 




LTl 


CM 





rA 


«— 


tA 




NO 


O 


C^ 


<r 


• 


LA 


On 


• 


(A 


•H 


•N 


r^ 




On 


<t 






CM 


CM 





On 
LA 

a 

CM 



00 

NO 

rA 

CM 



NO 



X 


















a. 


















<: 


















CD 
O 

LJ 


03 
03 
03 


NO 






<)• 


CO 


O 


c^ 


C 


• 


s? 




■ 


NO 


O 


LA 


■rH 


O 


r~ 


CM 


rA 


• 


NO 


LA 


JZ 


CM 


rA 


00 


T — 


rA 






Q 


CLJ 












NO 


CM 














CM 


CM 



LA 



• 




NO 




• 




• 


LA 




rA 




• 


rA 


CM 


a 




CM 




rA 


*s 




r> 




<d- 


rA 


r- 


o 
o 


T3 




■D 
03 




o 
rA 




CM 


■fefl- 








o 


rH 




4-) 


03 


>^ 








t-i 


• 




r— 


o 




03 


N 


.— 1 






03 


03 ^ 


u 




\ 






r—{ 


•H 


•H 






E 


-a 03 


>, 


c 


c 


CO 




O- 


CD 


E 






O 


c > 




u 


03 


C-i 




E 




CD 






O 


13 03 


Vt 


o 


u 


03 


U 


O 


>, 


l4- 




c^ 


c 


1 — 1 


03 


JZI 


■o 


03 


03 


o 


^H 




*^ 


r~~ 


•H 


en 


D3 




1 — 1 


>s 


03 




■H 


CJN 




ON 




03 >> 


CO 


03 


■H 




>-^ 


E 


r^ 


03 


r~ 


>> 


•H -U 




> 


SI 


<r 




o 


CO 


ON 


E 




f— I 


>— t U 


C 


■H 


o 


CM 


C 


o 


L*_ 


r— 


O 


C 


•H 


■H 03 


03 


+J 




1 


03 


^ 






o 


CO 


E 


E > 


•r-( 


03 


Cl_ 


LA 


•H 


o 


c 


C 


c 


•r4 


CD 


03 O 


■o 


c 


o 


^ — 


"O 


CO 


0! 


03 


■<-\ 


■o 


<4- 


L- Q- 


03 








03 




03 


03 




03 







lA 
00 



(4- 
CO 

u 
Q 



03 

E 
■D 
1— ( 
O 



o 

Q. 
03 

cn 

03 



o 

c 



c 

■rH 

u 

03 
CJ 

c 

03 
C_3 

c 
o 

03 
03 



o 

(L3 

to 



C_) 



o 



319 



conflicting values and a lack of familiarity with Western practices 
complicate their ease of entry into the health care delivery system. 

The full extent of our knowledge of the medical and emotional needs 
of these new immigrants is partially obscured by barriers of language and 
cultural distance; because they are a new population, our knowledge about 
their individual characteristics, patterns of utilization and variations 
among subgroups is fragmented. 

The refugee population is not evenly distributed across the country. 
Two-thirds of the population live in 40 American counties and secondary 
migration to areas of high Southeast Asian concentration after initial 
resettlement elsewhere is frequent. The population is young with 90 
percent of the population under 45 and 80 percent under 35.(57) Because of 
its age structure, contact with the medical care system often occurs 
through the need for obstetrical or pediatric care. 

Funding of medical services for refugees, particularly new arrivals, 
is through Title XIX State Medicaid programs, for refugees who meet State 
eligibility requirements, or through Refugee Medical Assistance, for 
refugees who meet State income requirements. The family composition 
requirement is waived for the first 18 months a refugee resides in the 
country, after which time he must qualify for assistance on the same basis 
as other indigent persons in the State. (58) 

Reliance on Medicaid for financing of medical services has been 
criticized because of its link to cash assistance. Fear of losing Medicaid 
benefits has been seen as an obstacle to employment, where medical 
insurance may not be available, and to self-sufficiency . (59) The 
limitation on benefits of State programs has led to difficulties for some 
immigrants, as in States where prescription drugs are not covered under 
Medicaid. 

A Rhode Island multi-year study has found that after initial 
resettlement, where refugee health screening is required, utilization of 
health services begins to lessen and decreases steadily over time. Most 
care sought is ambulatory with community health centers providing 72 
percent of outpatient care in Rhode Island. (60) Community health centers 
in general dominate in the delivery of health services to this population 
and thus provide the potential both for development of a data base on this 
population and a network for health education. (61 ) 

Lack of familiarity with Western medicine and language difficulties 
lead to problems in health seeking behaviors and compliance with physician 
instructions. Many are unfamiliar with the germ theory of disease and lack 
a surgical tradition. While such persons will seek a physician when they 
feel ill, they are not familiar with the idea of asymptomatic disease. 
Compared to other urban poor, there appears to be a delay in seeking care. 



320 



Language difficulties inhibit care. For example, difficulty in 
understanding a course of treatment, may cause treatment to stop when a 
symptom disappears. For those unfamiliar with Western life, accidents are 
frequent. (62) 

There is a strong tradition of self-treatment among Southeast Asians. 
The degree to which such practices as dermabrasive techniques and herbal 
medicines supplant Western medical care is not well known. (63) 

Utilization studies in different parts of the country suggest that 
the Hmong are low utilizers of health services while Vietnamese among 
Southeast Asian subgroups are high users of services. The latter' s 
prearrival experience with formal health services is cited as an 
explanation. (64) 



321 



Access to Health Care: Barriers 

Because this paper has presented only a recent snapshot of the status 
of various access-related indictors, it has not drawn attention to the 
progress that has been made in many of these measures in the past 20 years. 
These two decades have seen the halving of infant mortality rates, both for 
the general population and for blacks, the steady increase in percentage of 
women seeking early prenatal care, and some positive changes in the 
long-standing differential in health utilization statistics between high 
income and low income groups. But, as the foregoing discussion has shown, 
minority experience with many of the indicators relating to access to 
health care diverge from the majority experience. The remainder of this 
paper will comment on the obstacles which account for some of these 
differences. 

The recent past has seen changes in the nation's population — its age 
structure and population patterns, changes in its health care delivery 
system and changes in the environment in which the two interact. 
Nevertheless, the nature of barriers to health care access remain twofold 
in nature and unchanged from the past. 

The Robert Wood Johnson Foundation report found 12 percent of all 
Americans reporting access to health care difficult. This group had 
disproportionate representation of poor Blacks, Hispanics, the poorly 
insured and the unemployed. (65) Access to health care continues to depend 
on an absence of financial and structural obstacles to receiving care. In 
practical terms, financial access to health care translates into 
third-party insurance coverage. Despite the presence of Medicaid, there is 
a strong correlation between poverty and lack of insurance, with insurance 
coverage increasing with income. Uninsured persons of all ethnic and 
racial groups are less likely to have consulted a doctor in the past year 
than the insured(66) and use fewer ambulatory care services than the 
insured with minority uninsured persons using the fewest services. (67) 
Insurance is correlated to income: the lower incomes of minority groups 
make lack of insurance disproportionately a problem of minorities. 

Reliance of minority populations on the Medicaid program make these 
groups especially vulnerable to changes in eligibility or benefits 
associated with this program. The Robert Wood Johnson Foundation found 
that the cutbacks in public medical care service do not appear to result in 
a reduced volume of medical care received by minority groups, but it 
speculates that the lower levels of satisfaction reported by these groups 
suggest that these services are less effective in meeting their needs. (68) 

Structural barriers refer to such variables as physical distance, 
socio-cultural factors, language, and knowledge of the health care system 
which affect one's decision of when and how to seek care. Aday, Andersen 
and Fleming using 1976 data found urban Blacks more likely to contact a 
non-physician during an illness episode — 11 percent versus 7 percent for 



322 



the White population and Spanish heritage persons most likely to use home 
remedies in the treatment of illness — 21 percent as opposed to 12 percent 
for the White population. However, the degree to which such practices 
supplant conventional medical care is not understood. For example, 
although the use of curanderas, lay persons, for health advice is cited as 
a part of an Hispanic alternate health subculture, few studies report 
specific information on the number and location of these persons and the 
type of clientele they serve. (69) Studies have found that the degree of 
acculturation of Mexican-Americans, including knowledge of English, 
directly affects medical care utilization. (70) The extent to which being 
an illegal immigrant influences health seeking behavior has not been 
thoroughly examined. 

An array of divergent forces currently are working their influences 
on the nation's health care system. These include: 

the increasing physician supply 

the growth of new financial and organizational arrangements 

for the nation's physicians 

the growth of new technologies 

increased cost-consciousness among the purchases of health 

care and health insurance. This has led to a sharp rise in 

HMO membership, the restructuring of many health 

insurancepackages and increased awareness of the hidden costs of 

uncompensated care 

the growing role of proprietary organizations in the provision 

of health care services and a new business-oriented emphasis 

throughout the health care system. This has resulted in increased 

caution in determining the level and amount of indigent care 

institutions provide 

an increased emphasis on State-level approaches to solving 

public health policy problems. Some of these include: 

State/Local Programs of General Assistance 

A number of States and/or local governments have established 
programs to finance health services for poor individuals unable 
to qualify for Medicaid, under the heading of general assistance 
medical care. 

A few States have developed catastrophic health insurance 
programs. These include Alaska, Maine and Rhode Island. 

Revenue Pools 

A few States have initiated revenue pools which are used to 
finance indigent care. In New York, funds are derived from 
a surcharge levied on hospital reimbursement amounts paid to 
insurers. The pool funds are redistributed to individual 



323 



hospitals to offset charity care losses. In Florida, revenues 
are generated by a tax on each hospital's net revenues. The 
funds are used, in part, to pay for a major portion of the State 
contribution required to finance the medically needy program. 

Higher Levels of Medicaid Reimbursement for Certain Hospitals 

States were given increased flexibility under their Medicaid 
programs with the Omnibus Budget Reconciliation Act of 1981. 
New Medicaid approaches were required, however, to take into 
account the situation of hospitals serving a "disproportionate 
number" of low-income patients. These provisions provide States 
with an opportunity to direct funds to those hospitals which 
provide care to those unable to pay. Not many States have made 
extensive use of this approach to date. 

Direct Financial Support to Hospitals 

In addition to direct support for the operation of public 
hospitals, some States also have developed direct subsidy 
programs. Colorado has established a Medically Indigent 
Program in which direct payments are provided to hospitals to 
fund health care for uncovered individuals. California has 
established a County Health Services Fund which provides grant 
money and matching funds to counties for the provision of 
indigent care. 

Charting the effects of these changes upon the American people and 
their access to health care is a task for which existing information 
systems within government and the private sector are not well equipped. 
The limitations of existing data sources have been an underlying theme of 
the Secretary's Task Force, and these limitations are especially evident in 
the area of access to health services. There is a need to develop 
consensus as to the attributes of access which are most useful in making 
policy decisions and how these attributes should be monitored. The 
monitoring of access attributes should give particular attention to those 
population subgroups for which existing evidence suggests access to 
services is a problem. Greater attention needs to be paid to developing 
data collection strategies that are sensitive to the immediate needs of 
States and localities as they grapple with issues such as uncompensated 
care which are of particular interest to minority sub-groups. 
Modifications in national data collection efforts are possible which could 
make better data on racial and ethnic populations available. Also needed 
are improved methods for estimating changes in access in a rapidly changing 
health environment and means for relating access measures to health outcome 
measures. Better information about the national experience of access to 
health care and the experience of minority groups and subgroups will allow 
the better targeting of governmental and other interventions. 



324 



FOOTNOTES 

1. Aday, L., Andersen, R., and Fleming, G., Health Care in the United 
States; Equitable for Whom? , Sage Publications: New York, 1980. 

2. Ibid. 

3. Subcommittee on Cancer in Minorities Report , Volume III, Draft, 
February 1985. 

4. Ibid. 

5. Aday, L., Fleming, G., and Andersen, R., Access to Medical Care in 
the United States; Who Has It, Who Doesn't . Pluribus Press; 
Chicago, 1984. 

6. U.S. Bureau of the Census. Economic Characteristics of Households i n 
the United States: Fourth qTTarter 1983 , Series p. 70-83-4. 

7. Trevino, F., and Moss, A., "Health Insurance Coverage and Physician 
Visits among Hispanic and Non-Hispanic People," in Health United 
States, 1983 . DHHS Pub. No. (PHS) 84-1232. Public Health Service, 
Washington, United States Government Printing Office, 1983. 

8. "Health Insurance Coverage and Physician Visits among Hispanic and 
Non-Hispanic People." 

9. Ibid. 

10. Berk, M., and Wilensky, G., "Health Care of the Working Poor," 
"National Medical Care Expenditure Study, U.S. Department of 
Health and Human Services, Public Health Service, National 
Center for Health Services Research. 

11. Davis, K., and Rowland, D., "Uninsured and Underserved; Inequities in 
Health Care in the United States," Millbank Memorial Fund Quarterly/ 
Health and Society , 61:2. 1983. 

12. Access to Medical Care in the United States; Who Has it, Who 
Doesn't. 

13. Subcommittee on Cancer in Minorities Report . 

14. Ibid. 

15. Rossiter, L., and Wilensky, G. , "Out-of-Pocket Expenses for Personal 
Health Services," National Health Care Expenditure Survey, U.S. 
Department of Health and Human Services, Public Health Service, 
National Center for Health Services Research, Hyattsville, Maryland, 
1982. 



325 



16. Andersen, R., Giachello, A., and Aday, L., "Updating Access to Health 
Care among Hispanics," paper delivered at the American Public Health 
Association annual meeting, Anaheim, California, 1984. 

17. Kasper, J. and Barrish, G., "National Health Care Expenditures Study," 
Data Preview 12: Usual Sources of Medical Care and Their 
Characteristics DHHS Pub'. No. (PHS) 82-3324. U.S. Department of 
Health and Human Services. National Center for Health Services 
Research. 

18. Ibid. 

19. Montgomery, E., and Paranjpe, A., A Report Card on Health Maintenance 
Organizations; 1980-1984 , conducted for the Henry J. Kaiser family 
foundation by Louis Harris and Associates, Inc., undated. 

20. National Center for Health Statistics. Health United States, 1984 . 
DHHS Pub. No. (PHS) 85-1232. Public Health Service. Washington. 
U.S. Government Printing Office, Dec. 1984. 

21. Ibid. 

22. American Academy of Pediatrics. Trends in Pediatrician Participation 
in State Medicaid Programs . Chicago, III: American Academy of 
Pediatrics, March 1985. 

23. Children's Defense fund, American Children in Poverty . Washington, 
D.C.: Children's Defense Fund, 1984. 

24. Tenenbaum, J., "Health Care for the Medically Indigent" unpublished 
staff paper. National Health Planning Information Center. Office 
of Health Planning. Bureau of Health Maintenance Organizations and 
Resources Development, Health Resources and Services Administration, 
U.S. Public Health Service. 

25. Singer, J., Sulvetta, M., Solish, M. , and Beatrice, D., Uncompensated 
Care, Issues and Options , Draft, Waltham, MA: Health Policy Research 
Consortium, December 1984. 

26. Shanks, N., 12 Questions: What Legislators Need to Know about 
Uncompensated Hospital Care. Washington, D.C.: National Conference 
of State Legislators. Undated. 

27. Desonia, R., and King, K., Profiles of State Programs of Assistance 
to the Medically Indigent: A Report in Progress . Washington, D.C.: 
George Washington University Intergovernmental Health Policy Project, 
1984. 

28. Ibid 



326 



29. Uncompensated Care: Issues and Options . 

30. Ibid. 

31. Health Resources and Services Administration, Background Book , 
U.S. Department of Health and Human Services, Public Health 
Service. 

32. Ibid. 

33. Ibid. 

34. Trevino, P., "Health Indicators for Hispanic, Black and White 
Americans, Vital and Health Statistics , Series 10, No. 148 
DHHS Pub. No. (PHS) 84-1576, Public Health Service, Washington, 
United States Government Printing Office, September 1984. 

35. Ibid. 

36. Ibid. 

37. Subcommittee on Cancer in Minorities Report . 

38. Health United States, 1984. 

39. Ibid. p. 82 

40. "Health Indicators for Hispanic Black and White Americans." 

41. Ibid. 

42. U.S. Bureau of the Census, Statistical Abstract of the United States 
1985 (10th Edition). Washington, D.C., 1985. 

43. Indian Health Service, Chart Series Book , U.S. Department of Health 
and Human Services. Public Health Service, Health Resources and 
Services Administration, April 1985. 

44. Ibid. 

45. Gerzowski, M., and Adler, G., "Health Status of Native Americans," 
paper presented at the American Public Health Association annual 
meeting, Montreal, 1982. 

46. Ibid. 

47. HRSA Background Book 

48. IHS Chart Book 



327 



49. Ibid. 

50. Ibid. 

51. HRSA Background Book 

52. Statistical Abstract of the United States: 1985. 

53. Subcommittee on Cancer in Minorities Report 

54. Ibid. 

55. Kolter, M., Goldstein, M., and Lidsler, C, "Final Report of the 
Evaluability Assessment of the Centers for Disease Control 
Refugee Health Program." Macro Systems, September, 1984. 

56. Ibid. 

57. Moecke, M.A., "Caring for Southeast Asian Refugee Patients in the 
United States." American Journal of Public Health , 73:4, April 
1983. 

58. Forbes, S., "Medical Assistance for Refugees: Options for Change." 
Refugee Policy Group, Washington, D.C., November 1983. 

59. Ibid. 

60. August, Lynn Kao, "Health Service Utilization Patterns of Southeast 
Asian Refugees." Rhode Island Medicaid, Refugee Medical Assistance, 
Cranston, Rhode Island, April 1984. 

61. Interview with Dr. Robert Knouss, December 1984. 

62. "Caring for Southeast Asian Refugee Patients in the United States," 
p. 433. 

63. Ibid, p. 434-435. 

64. "Health Service Utilization Patterns of Southeast Asian Refugees." 
Also, Strand, P., and Clones, W., "Health Service Utilization by 
Indochinese Refugees." Medical Care , 21.11. 

65. Robert Wood Johnson Foundation. Special Report . Number One/1983. 
"Updated Report on Access to Health Care for the American People." 

66. "Health Insurance Coverage and Physician Visits among Hispanic and 
Non-Hispanic People. 

67. Ibid. 



328 



68. Access to Medical Care in the United States: Who Has It, Who 
Doesn't. 

69. Health Care in the United States: Equitable for Whom? 

70. Chesney, A., et al., "Barriers to Medical Care of Mexican Americans: 
The Role of Social Class, Acculturation and Social Isolation." 
Medical Care , 20:9, 1982. 

71 . Profiles of State Programs of Assistance to the medically Indigent: 
A Report in Progress 



329 



Health Education Among 
Minority Communities 

Report of the Working Group on 
Health Education 



The Task Force gratefully acknowledges Cheryl Daraberg, M.P.H., Office of 
Disease Prevention and Health Promotion, Department of Health and Human 
Services, for her valuable assistance in preparing this report. 



332 



Table of Contents 



I. Introduction 

II. Background 

A. Major Health Problems Facing Minorities 

B. Characteristics of Minority Populations that Impact on the 
Delivery of Health Education 

III. Developing a Health Education Strategy for a Minority Population 

A. Definition of Health Education 

B. Health Problems Amenable to Health Education 
Interventions 

C. Factors to Consider in Developing Strategies 

IV. Program Illustrations 

A. Primary Prevention Strategies with Low Income Hispanic 
Families 

B. Caide Su Corazon: Weight Reduction for Mexican 
Americans 

C. To Your Health - Living with Alcohol 

D. The California/Baja California Maternity Child Health 
Care Project 

E. Healthy Mothers, Healthy Babies Coalition 

F. Indian Health Service Diabetes Program 

V. Summary 

VI. Recommendations 

A. Information and Education 

B. Access and U .tilization 

C. Capacity Building in the Non-Federal Sector 

D. Financing Issues 

E. Health Professions Development 

F. Leadership, Work with Other Sectors 

G. Research Issues 
H. Data Issues 

VII. References 



333 



Health Education Among Minority Populations 



I. INTRODUCTION 

In the last century, dramatic changes have occurred in the leading 
causes of death and disability for most Americans. As documented 
in the 1979 Surgeon General's Report, Healthy People, infectious 
and communicable diseases no longer rank among the ten current 
leading causes of death. 1 Today, the leading killers are heart 
disease, cancer, stroke, accidents, influenza, diabetes, cirrhosis, 
suicide, and homicide. Many of the deaths attributable to these 
diseases are preventable, and thus unnecessary. 

Disease and disability are not events that are experienced equally 
by all individuals. For a given individual, the likelihood of 
developing a health problem depends on a variety of factors, for 
example, heredity, socioeconomic background, environment. 
Inadequacies in the health care system, and personal behaviors. 
Furthermore, the probability of experiencing ill health changes, 
depending upon individual experience with risk factors — behavioral 
and environmental influences that are capable of causing ill 
health, with or without previous disposition. 

Upon examination of the ten leading causes of death, controllable 
risk factors are identifiable for each. For example, heart disease 
is related to smoking, elevated serum cholesterol, diabetes, and 
obesity. In the areas of suicide and homicide, alcohol and stress 
are two prominent controllable risk factors. Some risk factors 
increase probabilities for several illnesses, such as cigarette 
smoking, poor dietary habits, alcohol, and severe emotional 
stress. It is interesting to note the dominance of lifestyle as a 
characteristic of each of these cross-cutting risk factors. 

For U.S. minority populations (Blacks, Hispanics, Native Americans, 
and Asian Pacific Islanders), as compared to the non-minority 
population (Whites) ,2 excess deaths are found in a number of the 
leading causes of death. The disparity in the Incidence and 
prevalence of certain health conditions for minority populations is 
a compelling reason to identify ways in which the health status of 
minorities can be improved. Because behavioral and environmental 
risk factors are associated with the causes of excess deaths among 
minorities, more work needs to be done in the area of health 
education, that is, for those components of the major health 
problems facing minorities that are amenable to health education 
efforts. Examples of these include the misuse of alcohol and 
drugs, use of tobacco, dietary habits, exercise, management of 
stress, adherence to medical regimens, and appropriate use of 
preventive services. (Table 1) 



334 



Table 1 



LEADING CAUSES OF DEATH FOR MINORITIES 
Cardiovascular Disease 



RISK FACTORS 



Smoking, high blood 

pressure, 

elevated serum 

cholesterol, 

obesity, diabetes, lack 

of exercise. 



Cancers 



Homicide, Suicide, and 

Unintentional Injuries 



Diabetes 

Infant Mortality 



Smoking, alcohol, solar 
radiation, worksite 
hazards , environmental 
contaminants, diet, 
infectious agents. 

Alcohol or drug misuse, 
stress , handgun 
availability. 

Obesity. 

Low birth weight, 
maternal smoking, 
nutrition, stress, 
trimester of first 
care, age, marital 
status. 



Cirrhosis of Liver 



Alcohol. 



For the four minority groups identified, as for any group, health 
education interventions are directed at improving the awareness of 
individuals and communities about controllable risk factors 
associated with the causes of excess death and disability. 
Differences in health status underscore the importance of providing 
health education to minority populations but consensus has not been 
reached on how to develop health education programs and strategies, 
how to affect change, and how to disseminate these strategies. It 
is somewhat unrealistic to expect consensus, however, due to the 
constant shifts occurring in these populations as they become more 
"main stream," and because of the existence of very different 
segments within these minority groups. 

The purpose of this paper is to describe how health education can 
be used to address the health problems of minority groups. The 
paper starts with a brief overview of the major health problems 
facing Blacks, Hispanics, Native Americans, and Asian Pacific 
Islanders and a discussion of minority group characteristics. 
Next, the paper describes the potential for health education to 
impact positively on minority health status. Following this 



335 



section is a discussion of relevant factors to consider when 
developing a health education program. A sampling is then provided 
of health education interventions that have been undertaken to 
illustrate ways in which minority populations can be reached. In 
the final portion of this paper, recommendations for future health 
education activities for minority populations are presented. 

II. BACKGROUND 

A. Major Health Problems Facing Minorities 

Based on National Center for Health Statistics (NCHS) 
death certificate data for 1979 and 1980, areas of excess 
deaths which contribute to 80% of the disparity in health 
status between non-minorities (Whites) and minority 
populations (Black, Hispanic, Native American, and Asian 
Pacific) were calculated as the data were available. 2 
Six major problem areas were identified: cardiovascular 
disease, cancer, infant mortality, diabetes, violence 
(homicide, suicide and unintentional injuries), and 
alcohol and drug misuse. It is worth noting that each of 
the problem areas for the four minority groups have 
elements amenable to health promotion and disease 
prevention interventions. 

Blacks 

Using 1979-1980 data, the percent of excess deaths by 
leading disease category for Black males under 45 
years of age was 36.6% for homicide, 14.3% for infant 
mortality, 9.7% for heart disease, 7.8% for 
accidents, 4.9% for cirrhosis, and 3.0% for 
pneumonia, as compared to White males under 45 years 
of age. For Black males under 70 years of age, 
compared to their White male counterparts, the 
percent of excess deaths within each disease category 
was 18.7% for homicide, 16.1% for cancer, 15.5% for 
heart disease, 8.1% for cerebrovascular diseases, 
6.7% for accidents, 5.9% for infant mortality, 4.0% 
for cirrhosis, and 3.4% for pneumonia. 

For Black females in this same time period, the 
percent of excess deaths for those under 45 years of 
age by disease category was 21.6% for infant 
mortality, 13.9% for homicide, 13.1% for heart 
disease, 5.1% for cirrhosis, 5.0% for cancer, 4.3% 
for cardiovascular, 4.2% for accidents, and 3.3% for 
pneumonia, as compared to White females under 45 
years of age. The percent of excess deaths for Black 
females under 75 years of age was 30.5% for heart 
disease, 10.5% for cardiovascular, 9.5% for cancer, 
7.5% for infant mortality, 5.9% for homicide, 5.0% 
for diabetes, 3.4% for cirrhosis, and 2.6% for 
nephrltis/nephrosis. 



336 



Hispanics 

Until recently, national data on Hispanic health 
status indicators did not exist; therefore, 
definitive statements could not be made about the 
problem areas which contribute to the disparity in 
health status between Hispanics and non-minorities 
(Anglos). Action was taken to remedy this problem in 
the last several years, and national data on 
Hispanics will be forthcoming. While selected State 
Hispanic data exist, there are significant problems 
with how the data were gathered. It is impossible to 
use the State data, with the inconsistencies in the 
methods of collection, to make generalizations by 
region or by Hispanic group. 

It is known that many Hispanics in the United States 
are living below the poverty line, and that as a 
group Hispanics have much lower levels of education 
than do non-minorities (Anglos). Both of these 
factors ultimately impact on the health status of 
Hispanics. Some evidence indicates that gall bladder 
disease, obesity, diabetes, cardiovascular disease, 
and tuberculosis are among the more significant 
health problems facing individuals of Hispanic origin. 

Native Americans 

The percent of excess deaths for Native American 
males under 45 years of age, by leading disease 
category for the period 1979-80, was 49.5% for 
accidents, 10.7% for cirrhosis, 9.5% for homicide, 
8.1% for suicide, 3.3% for heart disease, and 3.0% 
for pneumonia, as compared to Whites. For Native 
American males under 70 years of age, the percent of 
excess deaths within each disease category was 56.7% 
for accidents, 17.8% for cirrhosis, 10.4% for 
homicide, 7.3% for suicide, 4.9% for pneumonia, 3.9% 
for diabetes, and 2.9% for heart disease. 

For Native American females under 45 years of age, 
the percent of excess deaths within each disease 
category was 38.9% for accidents, 20.0% for 
cirrhosis, 6.8% for homicide, 4.5% for heart disease, 
3.8% for diabetes, and 3.6% for pneumonia, as 
compared to Whites. The percent of excess deaths for 
Native American females under 70 years of age was 
41.0% for accidents, 29.7% for cirrhosis, 11.1% for 
diabetes, 7.3% for homicide, 5.7% for 
nephritis/nephrosis, 4.3% for pneumonia, and 3.9% for 
heart disease. 



337 



Asian Pacific Islanders 

Within the major disease categories, the incidence of 
disease among the Asian population, for males and 
females, is lower than in the non-minority 
population. However, the risk for suicide in Chinese 
females rises considerably after age 45 and increases 
with advancing age. Areas of greatest concern where 
incidence is higher for Asian Pacifies, particularly 
for the most recent immigrants and refugees, are 
tuburculosis and hepatitis. 

B. Characteristics of Minority Populations that Impact on the 
Delivery of Health Education 



Blacks 



Demographics 

The Black population is the largest minority 
group living in the United States, approximately 
26.5 million in number. The majority of Blacks 
live in urban areas. As is true with the other 
minority populations discussed in this paper. 
Blacks are not homogeneous as a group. Blacks 
exhibit great variation in educational levels, 
socioeconomic status, and religion. While some 
similarities exist, differences among Blacks, 
region by region, group by group, person by 
person, must also be considered when designing a 
program. 

Health Beliefs 

Although a reliable estimate for the prevalence 
of folk or traditional health beliefs and health 
care providers in the Black population does not 
exist, it is important to be cognizant of the 
beliefs and practices which exist since they may 
ultimately Impact on a health education 
intervention. It is believed such traditional 
beliefs and practices are more prevalent in 
individuals who have less access to mainstream 
health care, such as older, lower income, rural 
dwelling individuals. 3 Blacks living in urban 
settings, as the majority do, tend to use the 
mainstream system as their first choice of care, 
but some may still hold traditional beliefs. 
Few studies to date have examined health beliefs 
in the Black population. 



338 



Sources of Health Information 

Blacks, in general, tend to have well-organized 
communities which play an important and active 
role in health and social matters. In many 
Black communities, for example, the church 
serves as a powerful and influential institution 
around which the community may organize. The 
church often is the primary institution of 
social support and social control, and 
significantly influences the norms guiding the 
lives of church members."^ Because ministers, 
deacons, ushers and other church personnel are 
seen as influential, their potential for serving 
as health education advocates is great. The 
importance of the church in Black communities 
should not be underestimated or ignored in 
planning a health education strategy. The 
well-organized and highly credible church 
organization could be effectively used to 
screen, educate, and follow-up on individuals in 
the community who are at increased risk of 
certain diseases. 

The extended family, traditionally, has been a 
primary source of health information for Blacks, 
in part, due to their limited access in the past 
to the mainstream health care system. The 
important role of the Black family as providers 
of counsel and support continues , and may 
provide an effective means for conveying health 
information. Because of the high value placed 
upon families in Black communities, family 
members could be used to carry health messages 
that stress the importance of the individual 
family member's health for the security and 
well-being of the entire family. Messages 
incorporating the theme "do it for your loved 
ones" have been quite effective in reaching 
members of the Black community. 3 

In a study completed in 1982 by Juarez and 
Associates, Black pregnant women and recent 
mothers named the physician as the best and most 
credible source of health information. 5 xhe 
study also found that it did not make a 
difference whether the information was given 
verbally or handed to the women. Nurses were 
not seen as credible sources of health 
information by these women because they were 
more often associated with the negative aspects 
of clinic care. Women, Infants, and Children 



339 



(WIC) program nutritionists were also cited as a 
favored source of health information. 

In another study, Gombeski e^ al. collected data 
on Whites, Blacks, and Mexican Americans to 
determine the media habits of Houston-area 
adults. 6 According to the results of the 
community survey. Blacks identified physicians 
as their primary source of health information, 
with television as the second major source. 
Whites, on the other hand, selected newspapers 
as their primary source and physicians as the 
second major source. In terms of the most 
credible source of health information, the 
physician was overwhelmingly selected by Blacks, 
Whites, and Mexican Americans. These data 
indicate that the most available source of 
information may not necessarily be the most 
credible source (physician). Because physicians 
are reported as the most credible source of 
health information and education, they should be 
encouraged to counsel their patients about 
measures to promote health and prevent disease. 

To reach Blacks effectively as a target group, 
health education for Black communities should 
take place in settings relevant to Blacks, such 
as in the neighborhood, at the worksite, through 
the media, in the schools, churches, and other 
community organizations that serve Blacks. 4 
Because of the important role of the community 
in Black life, active participation by the 
community in program development is an essential 
aspect of a successful intervention. 

In addition, health education programs in the 
Black community require an understanding of the 
relationship between the environment, lifestyle, 
and the improvement of the quality of Black 
health. 4 Programs may need to address other 
priorities for Black communities such as 
education, job training, and child care 
programs, in addition to addressing the health 
problem(s). All the major determinants of 
health must be acknowledged in the design of 
health education interventions. Programs cannot 
be problem oriented alone, but must be addressed 
from a quality of life context. As such, health 
educators may have to assume advocacy roles to 
assist Black communities in addressing political 
and social issues that either impact on health 
status or are of greater priority than the 
identified health problem alone. 



340 



Hispanics 

The Hispanic population represents a "mosaic" of 
individuals and groups of individuals. ^ Because 
Hispanics derive their cultural make-up from a 
variety of countries of origin (Mexico, Cuba, Puerto 
Rico, Central and South America), they cannot be 
classified as a homogenous population. Furthermore, 
as a population, Hispanics vary greatly in their 
level of educational attainment, acculturation, and 
socioeconomic development. While Hispanics differ 
among themselves as a group and are in a constant 
state of change, good generic Hispanic health 
education programs are possible if carefully 
developed and tested. Only when a particular ethnic 
subgroup characteristic affects the behavior in 
question, such as their media use or access to care, 
should health education programs target subgroups 
separately. While the Spanish language is the 
primary cohesive factor binding Hispanics together, 
their cultural, economic, and educational diversity 
dictates that no single health education strategy is 
likely to suit the hetrogeneous needs of this 
population. 8 

• Demographics 

An important characteristic of the Hispanic 
population in comparison to non-minorities 
(Anglos) is its much younger age structure. The 
Hispanic population has a median age of 23, with 
one-third of its population under the age of 
15. Currently, the Hispanic population is the 
fastest growing minority group in the United 
States, comprised of approximately 14.6 million 
people. 3 Of this total, 60% of Hispanics are 
of Mexican origin. Most Hispanics live in urban 
areas. 

As a whole, the level of educational attainment 
for Hispanics remains far below that of 
non-minorities. Even though upward mobility in 
educational attainment exists for subsequent 
generations of original immigrants, the rate 
lags far behind those of European immigrants. 
This disparity in educational attainment 
contributes to Hispanics having higher rates of 
unemployment and greater numbers of individuals 
living below the poverty line than for 
non-minorities. Among all Hispanics, Mexican 
Americans and Puerto Ricans tend to fare the 
worst economically. 

341 



Hispanic families are typically larger than 
non-Hispanic families, with Mexican Americans 
having the largest families among all 
Hispanics. Due to their generally lower 
economic status and large family size, Hispanics 
have high rates of overcrowding. Overall, 
Hispanic families are experiencing shifts in 
family structure similar to the general 
population; for instance, Hispanics have an 
increasing number of female-headed households 
and single-member households. 

Family Involvement in Health Maintensnce 

The role of the extended family as an available 
source of health information for Hispanics is an 
important issue to consider. Hispanics tend to 
have strong family support networks. There is a 
tendency for lower socioeconomic and culturally 
diverse minorities to be more ethnocentric and 
isolated from mainstream health care. As a 
result, they may rely heavily on family members 
for support and information. Roles in Hispanic 
families are clearly defined: younger members 
defer to elders , and men are considered the 
authority figures in their families. 

A strong family structure can serve to either 
encourage or discourage the seeking of health 
care and the use of health information by 
individuals. 10 Depending on the level of 
acculturation and socioeconomic status, Hispanic 
families may serve to reinforce traditional 
beliefs about health, or they may maintain 
barriers against participation in mainstream 
health care. 3 por example, one study found 
that young pregnant Hispanic women were largely 
influenced by their mothers and grandmothers, 
who provided them with less accurate health 
information than their doctor or nurse. 5 

For many Hispanics, illness is a family affair 
in that the extended family is involved in 
making decisions about the course of treatment. 
Hispanic families traditionally share benefits 
among members and avoid seeking outside help, 
since loyalty to the family overrides individual 
Interests. 3 Thus, when conveying health 
education messages, it is essential that the 
Hispanic family be treated as a unit rather than 
as individual members in isolation in order to 
avoid possible barriers and to enhance the 



342 



educational effort. In addition, lines of 
authority and decision making should be 
respected and attended to when designing or 
implementing health education interventions 
among Hispanic populations. 

Perceptions of Health and Disease 

The health beliefs of Hispanics represent a 
mixture of traditional health beliefs and 
"scientific" or mainstream medical beliefs. 
According to da Silva, some "Hispanics define 
health as the ability to work, resulting from 
good luck or good behavior or from a gift of 
God. Illness is seen as the presence of 
symptoms and is often accepted 
fatalistically. "9 it is also worth noting 
that some Hispanic complaints have no convenient 
reference point in the lexicon of Anglo 
medicine, such as empacho (indigestion), caida 
de la mallera (sunken fontanel), mal ojo (evil 
eye), and mal aire (bad air). 

Because Hispanics may have traditional 
perceptions of health and illness that do not 
necessarily coincide with contemporary medical 
beliefs and practices, the dissemination of 
health information with scientific deriviations 
may create conflicts for many Hispanics and may 
not be received or used as intended. Should 
traditional beliefs be determined to play a role 
in an Hispanic community, messages should be 
tailored within existing belief systems whenever 
possible rather than attempting to completely 
restructure beliefs and practices. 

Personal Characteristics 

Several individual behavioral factors common 
among Hispanics can play a key role in both the 
diffusion of health information and the seeking 
of health care. Among some groups of Hispanic 
adult males, "machismo" attitudes are 
prevalent. 3, 9 xhe degree of importance placed 
on exhibiting a macho attitude will impact on 
whether or not an Hispanic male will use the 
health system himself or whether his family will 
use mainstream health care services. Machismo 
also is closely associated with one's perceived 
susceptibility to an illness, and will 
ultimately affect the degree and ease to which 
behavior change through health education is 
possible among Hispanics who possess this 
personal characteristic. 

343 



Another important characteristic seen among 
Hispanic women is modesty. 9 por Hispanic 
women, modesty may be a reaction to or a result 
of male dominance; in some cases a husband may 
not want a male to treat his wife for a health 
problem. It is thus possible that modesty may 
prevent some Hispanic women from seeking health 
care, including preventive services and health 
education. 

Determining whether such characteristics as 
these exist within an Hispanic community that 
has been identifed as the target for an 
intervention allows for the design of the 
appropriate health education/ information 
strategy. 

Sources of Health Information 

A study in San Antonio, Texas, which examined 
the media consumption habits of individuals of 
Mexican origin, found that no significant 
difference exists between urban Mexicans and 
non-minorities (Anglos) in the average number of 
hours spent each week watching television. 6 
However, Mexicans were more likely than 
non-minorities to watch television on Sundays 
and to watch late evening news broadcasts. 
Particularly noteworthy is the finding that 
information presented on national news programs 
was viewed as more credible by Mexicans than by 
non-minorities. In the area of radio, Mexicans 
were more likely to listen to radio than 
non-minorities. A main reason cited by Mexicans 
for station preference was the desire to hear 
Spanish music, since music is an integral part 
of Hispanic life. With respect to newspapers, 
Mexicans were more likely than non-minorities to 
read the sports page and special advertising 
supplements, but were less likely than 
non-minorities to read the editorial section. 
Mexicans in San Antonio were also more likely to 
subscribe to weekday afternoon and Saturday and 
Sunday morning editions of a newspaper. 

The findings from this study indicate that urban 
Mexicans are sufficiently different from 
non-minorities to merit a special communication 
effort to reach them effectively. Individuals 
and organizations wishing to convey health 
information to individuals of Mexican background 
and other Hispanics need to consider carefully 



344 



Table 2 



preferences in the type of media used as well as 
how that media source is used. 

In another study, the Baylor College of 
Medicine, National Heart and Blood Vessel 
Research and Demonstration Center, completed an 
in-house community survey of Houston area 
adults." The study examined the sources of 
health information for Whites, Blacks, and 
Mexican Americans. As indicated in Table 2, 
physicians represent a major source of health 
information for Mexicans, both male and female. 



MAJOR SOURCE OF HEALTH INFORMATION 
IDENTIFIED BY ETHNICITY OF RESPONDENT 









Mexican 




Source of Health 


White 


Black 


American 


TOTAL 


Information 


N=1604 


N=547 


N=170 


N=2,340 


Doctor 


21.8 


37.3 


35.0 


26.4 


Newspaper 


27.8 


9.3 


13.5 


22.3 


Magazine 


12.0 


7.0 


5.0 


10.3 


Television 


16.0 


25.0 


24.1 


18.7 


Radio 


3.0 


3.6 


3.5 


3.0 


All Other Sources; 










Don't Know 


19.4 


17.8 


18.9 


19.3 



TOTAL 
Table 3 



100.0 



100.0 



100.0 



100.0 



MOST ACCURATE SOURCE OF HEALTH INFORMATION 
IDENTIFIED BY ETHNICITY OF RESPONDENT* 



Most Accurate White 
Source of N=1604 
Health Information % 



Black 

N=547 
% 



Mexican 




American 


TOTAL 


N= 170 
% 


N=2340 
% 


64.1 


69.1 


3.5 


4.1 


6.5 


8.2 


5.3 


6.2 


0.0 


0.3 



Doctor 


68.1 


Newspaper 


4.7 


Magazine 


9.5 


Television 


5.4 


Radio 


0.2 


All Other Sources; 




Don't Know 


12.7 



73.5 
2.4 
4.9 
8.6 
0.9 

9.7 



20.6 



12.1 



TOTAL 



100.00 



100.0 



100.0 



100.0 



345 



In addition, physicians were viewed as the most 
accurate source of health information for Mexicans 
(Table 3). However, many poor Hispanics have little 
contact with health professionals due to possible 
cultural and language barriers. As a result, several 
studies have found that many Hispanics rely on the 
mass media as a source of health information. 

• Language 

For Hispanics, Spanish is still the dominant and 
preferred language. 6 However, bilingualism is 
growing, with a reported 78% of Hispanics 
stating they either speak or understand 
English. Generally, Spanish is the preferred 
means of communication. Exceptions to this 
generalization are among those individuals who 
are younger and more educated, who may have only 
learned to read English. Individuals who only 
speak Spanish tend to be concentrated into 
several groups — recent arrivals, the very young, 
the very old, and those unable to read any 
language . 

Native Americans 

Native Americans have a number of health problems 
that warrant health education interventions, 
especially in the areas of alcohol abuse and fetal 
alcohol syndrome, unintentional injuries, coronary 
heart disease, gall bladder cancer, diabetes, and 
obesity. 11 However, in attempting to design an 
appropriate intervention for Native Americans, an 
educator should be aware of the tremendous diversity 
that exists within this small population. 

• Demographics 

According to the 1980 census, there are 
approximately 1.5 million Native Americans 
residing in the United States. This population 
consists of American Indians, Eskimos, and 
Aleutian Islanders. Native Americans represent 
the smallest minority population living in the 
U.S. and are concentrated primarily in the 
Southwestern section of the country. 3 Over 
50% of this population resides in rural areas. 



346 



Native Americans are separated into 212 tribes, 
with the five largest tribes being Navajo, 
Cherokee, Sioux, Chippewa, and Pueblo. What is 
important from an educational perspective is 
that each tribe possesses its own 
dialect/language, philosophy, customs, and 
structured tribal government. This diversity 
further complicates the design of a health 
education intervention, since no single design 
will suit the needs of all Native Americans. 
While common elements can be found among Native 
Americans, educators and communicators need to 
address the particular tribal characteristics 
that will ultimately impact on the development 
and implementation of an effective intervention. 

Cultural Use of Tobacco 

A substantial problem exists when interventions 
are targeted at health practices or behaviors 
that are affected by the cultural milieu. For 
example, tobacco use is interwoven into Indian 
cultural and religious practices, 12 therefore 
the prevention of smoking becomes problemmatic. 
Traditionally, tobacco has been viewed as a 
sacred substance to be used ceremonially. 
Smoking education for this population, while not 
a high priority from a Native American 
standpoint, needs to focus on improving the 
awareness of the health hazards associated with 
the habitual use of cigarettes. Existing 
organizational structures such as Indian Health 
Service clinics. Tribal Boarding Schools, and 
other Indian community organizations should be 
used to disseminate health information on 
smoking. Messages must be carefully tailored 
around existing cultural and religious beliefs. 

Sources of Health Information 

American Indians have three important sources of 
health information in addition to mainstream 
health care providers/physicians. 3 One of 
those sources is the extended family, which is a 
source of health information for other minority 
groups as well. This is, undoubtedly, due to 
the perception that family members are more 
likely to be sympathetic about and understand 
the health problem than are outside mainstream 
health care providers. The extended family and 
clan are integral to most tribes' social 



347 



structures. Households are frequently composed 
of the nuclear family plus members from three or 
four generations. As a result, a health 
education intervention could be enhanced by the 
inclusion of more than just the individual or 
nuclear family in the diffusion effort. 

Another source of health Information Is 
traditional healers/medicine people. In 
traditional Indian cultures, illness is viewed 
as a state of disharmony or imbalance. 
Traditional healing ceremonies conducted by 
medicine people are used to correct the 
imbalances causing the disease. Indians tend to 
use medicine people, in addition to mainstream 
health care providers, because of existing 
traditional health beliefs. Educators and 
communicators need to assess the extent to which 
this alternative health care system is used by a 
particular Indian community to determine how and 
to what degree such beliefs and practices will 
affect the health education effort. It is 
always important to remember, however, that 
changes in health practices will be more 
acceptable to the Indian community provided they 
do not challenge the existing culture and 
beliefs. 

A third source of health information is the 
Community Health Representative (CHR) , within 
the Indian Health Service; it serves as a 
liaison between the tribal community and 
mainstream health care providers. These 
representatives are community people who are 
trained to assist Indian communities in 
obtaining services, both existing and new; to 
provide basic health instruction to Indians in 
their homes and the community; and to coordinate 
the provision of comprehensive health services 
to the Indian community. CHRs are important to 
the success of health education efforts since 
they serve as a vital source of information to 
the community. CHRs are highly credible sources 
of health information and could serve as 
effective agents of change because they are 
known members from within the community who have 
been trained to provide health education 
services. It should be noted, however, that 
each tribe has a slightly different CHR program 
that reflects the particular tribe's health 
needs and priorities. 



348 



For Indians, the presence of an individual with 
the same ethnic background assists in the 
diffusion of health information. 2 This is due 
to the fact that members from within the Indian 
community have status and credibility as sources 
of information. Indians generally prefer 
practitioners of Native American background, 
since they possess a cultural understanding and 
sensitivity toward Native Americans. 

Urban versus Rural Environment 

American Indians are provided with health 
services and health education through the 
programs of the Indian Health Service (IHS). 
Primarily, the IHS serves Indian communities in 
rural and isolated environments. For those 
individuals living on reservations, tribal 
councils, tribal newspapers, alcoholism 
counselors, and community organizations provide 
good mechanisms for information dissemination 
and educational efforts. Within Indian 
communities there is a great deal of movement 
between the reservation and urban centers, 
because of the need to find employment. Indians 
living in urban areas may have greater access 
and exposure to television programming and 
health messages aimed at the general population 
than do rural Indians ; however , they lack the 
support and diffusion mechanisms provided by 
Tribal Councils and the Indian Health Service. 
There also is a lack of Indian-directed radio or 
television programming for urban dwellers. 

Several potential avenues exist for reaching 
urban Indians with health education messages. 
For example, urban Indians tend to cluster in 
neighborhoods where they form clubs, community 
centers, and churches, through which health 
information could be diffused. 

Message Content 

Because great value is placed on the extended 
family as well as on ties to former ancestors 
among Native Americans, messages should be 
designed to highlight aspects of Indian life and 
culture that can be used positively to influence 
health behavior. Messages that stress the 
importance of "keeping healthy for your 
children" or "eating and keeping fit as your 
ancestors did" could be effectively employed. 3 



349 



Asian Pacific Islanders 

The Asian Pacific population is highly diverse in 
terms of cultures, customs, and languages, and thus 
presents many challenges to educators/ communicators 
seeking to disseminate health information. Because 
no national data exist that break out disease 
pravalence and incidence by country of origin for 
Asians, the health educator must assess the needs of 
each target population on a community level. 

• Demographics 

The Asian population in the United States has 
grown significantly in the last ten years, 
partially due to an influx of refugees from 
Cambodia, Laos, and Vietnam. In 1980, the 
Census Bureau reported 3.5 million Asian 
Pacifies living in the United States, the 
majority of whom reside in urban areas. This 
minority group consists of individuals from 
Japan, China, Samoa, Korea, Thailand, Burma, 
Hawaii, Vietnam, Laos, Cambodia, and the 
Philippines. With such a large number of 
countries of origin, the Asian Pacific 
population exhibits tremendous diversity in 
customs, languages, religions, educational 
levels, level of acculturation, and 
socioeconomic status. The range extends from 
poor and uneducated refugees to more highly 
educated and financially sound immigrants. The 
newest wave of Asian immigrants have come 
primarily from Vietnam, Cambodia, Laos, and 
Burma. These immigrants are more prone to 
culture shock than earlier immigrants from Japan 
and China because the majority have not been 
exposed to an industrial/western way of life. 
Generally, the priorities for the newest 
arrivals are securing housing, jobs, and 
language skills. Often, health is not a high 
priority for these individuals who may be 
experiencing a great deal of frustration in 
adjusting to American life. 

• Family Structure 

For the most newly arrived Asian immigrants and 
their families, the effects of urbanization, 
role changes in the family, cultural conflicts, 
and acculturation create feelings of relative 
powerlessness and social isolation within 



350 



American society. As Asian families become more 
acculturated, with subsequent generations, their 
family structure becomes more like the majority 
population. 3 Newly arrived families are more 
likely to live together for both economic and 
social support reasons, while those who are 
better acculturated to American life may live 
alone, may divorce, or may not live with 
elders. However, family influence upon members 
remains strong even if the family does not 
reside together. Males tend to still serve as 
the heads of the household, and age, experience 
and seniority are respected. Health education 
interventions would, therefore, need to consider 
the role that family influence plays in 
communicating health information. It might be 
best to design programs that reach family units 
as opposed to individuals, and to acknowledge 
the lines of authority and decision-making 
within Asian families. Additionally, because 
the church plays a significant role in most 
Asian communities, its potential is great for 
reaching families with health information. 
Also, messages with a "love of family/ children" 
theme would tend to be appropriate for Asians, 
given the importance of the family network. 3 

Community Boundaries 

Subgroup communities of Asian Pacifies are 
frequently self-contained, as in the case of 
Chinatowns and Koreatowns. Since residents of 
these communities generally do not travel beyond 
community boundaries for services and because 
they often encounter cultural and language 
barriers, health education/ information services 
need to be brought to the target community. In 
addition, Asians who require health 
education/ information services in their native 
language usually find culturally relevant and 
language-appropriate materials to be virtually 
nonexistent. 

To increase the likelihood of success, 
education/ information efforts must involve 
community outreach to develop suitable programs 
and materials. By working with existing 
community organizations, such as ethnic 
churches, associations of business owners, 
community centers and the Asian Pacific caucus, 
programs can be appropriately designed and 
disseminated at the community level. 



351 



Health Beliefs 

Traditional health beliefs and practices may 
exist for many Asian Pacifies, especially for 
those who are first generation families, 
elderly, and live in highly concentrated and 
segregated Asian communities. It is likely that 
traditional beliefs and practices are most 
prevalent among recent Indochinese immigrants 
and least prevalent among Japanese Americans. 
One should note that it is extremely difficult 
to generalize about the different Asian Pacific 
ethnic cultures; therefore, it is essential for 
the health educator/ communicator to assess the 
prevalence of folk beliefs and practices, and to 
then determine how such beliefs impact on the 
design and delivery of health education 
interventions . 

Sources of Health Information 

Asian Pacific populations have three basic 
sources of health information: (1) the 
mainstream system; (2) the folk/ traditional 
system; and (3) unlicensed medical professionals 
in the community. 3 The decision to seek 
health care or health information depends on the 
individual and his/her level of acculturation, 
the ailement , and the availability and 
acceptability of health care providers. More 
often than not, the mainstream system is used. 
Prior to designing an educational/informational 
intervention for the Asian population, a careful 
assessment of recipient needs, beliefs, and 
sources of health information is required. 

A paucity of data exists in the area of what 
sources of health information are used by Asian 
Pacifies, which precludes one from making 
definitive statements about the best course to 
pursue; however, it is possible to speculate 
about how to develop and place health messages 
for Asians based on the limited information that 
is available. For the most recent Asian Pacific 
immigrants, effective messages may have to be 
bilingual. 13 This may require that mainstream 
providers of health education and care provide 
interpreters to ensure adequate communication 
with members of the target population. Another 
potential mechanism for disseminating health 
information is through the print media, such as 



352 



(1) community and national ethnic newspapers, 

(2) pamphlets distributed In ethnic food stores 
and at community centers, and (3) newsletters 
that are distributed and read weekly at church. 
With respect to the use of television, stations 
that provide non-English language programming 
should be considered as another viable avenue 
for the diffusion of health Information to Asian 
Pacifies. 

• Cultural Characteristics 

The desire for independence and self-sufficiency 
among Asian Pacifies is strong, and in general 
there exists a substantial fear of dependency. 
Because this tends to be true, health promotion 
and disease prevention messages are likely to be 
more effective if they convey a need to take 
care of one's health to avoid a problem, and 
thus avoid dependency upon someone else.^ 

Another common characteristic found among Asian 
Pacifies is their great sense of being in 
control of or being able to control their own 
bodies and destinies. Whenever possible, this 
trait should be capitalized on to convey health 
messages; that is, the educator should emphasize 
the role the individual can play in adopting or 
changing behaviors to control or improve health 
status. 

III. DEVELOPING A HEALTH EDUCATION STRATEGY FOR A MINORITY 
POPULATION 

A. Definition of Health Education 

Health education for any population focuses on improving 
the awareness of individuals or communities about 
lifestyle/behavioral issues that Impact on health status. 
According to Hochbaum, 

...it is true that the more one knows, the more likely 
that person is to act in a healthy manner. . .knowledge may 
even motivate one to engage in sound health practices, but 
knowledge alone is not enough to ensure that the 
individual will act in a healthy manner or choose positive 
health behavior s.l^ 

Hochbaum points out that an individual's actions are 
influenced in many ways, by such factors as competing 
priorities and incompatible messages. 



353 



Another definition, provided by Green et al. , states that 
health education consists of "...any combination of 
learning experiences designed to facilitate voluntary 
adaptations of behavior conducive to health. "15 
Furthermore, Green ascertains that while the terms health 
information and health education are used inter changably, 
the two processes are distinct entities. For Green, 

Health education. . .refers to strategies or learning 
experiences designed to bring about voluntary adjustment 
of behavior conducive to health. . .Information may be one 
element of a strategy, but most behavioral change requires 
more than health information. 16 

Health information, then, is often neither a necessary nor 
a sufficient component of effective health education 
interventions. Clearly, individuals who are the target of 
health education will require some minimum level of 
information to engage in voluntary behavior change; 
however, health education ideally assists the individual 
in making healthful decisions, modifying risk, factors, 
adhering to therapeutic regimens, or resisting pressures 
to engage in harmful practices. 

In summary, minority health education interventions seek 
to facilitate coimnunity and individual measures that can 
foster the development of lifestyles to maintain and 
enhance the state of health and well-being, as well as to 
increase public and professional awareness of risk factors 
that impact on minority health status. Given the breadth 
of such interventions, it is useful to keep in mind that 
health education interventions can occur in a multitude of 
settings — for instance, in schools, worksites, and 
coimnunities. 

B. Health Problems Amenable to Health Education Interventions 

Among the most important variables that ultimately affect the 
outcomes of health education interventions are the following: 

• the nature of the target population; 

• the types of interventions available; 

• the kinds of health-related outcomes that can be 
expected from the intervention; 

• the kinds of health-related outcomes that should be 
of high priority; and 

• the outcomes other than health-related that may be 
expected. 

354 



All of these issues are critical to the successful planning, 
adoption, and implementation of health education strategies for 
minority populations. In order to address the particular needs of 
the populations, the educator must know the critical problems, the 
likely delivery mechanisms, and strategies for implementation. He 
must recognize the types of interventions that are available given 
a unique set of needs and population characteristics (e.g., 
interventions may be regulatory or legislative, educational 
products, organizational structures/ change, personnel training, 
technical assistance, or personnel assistance). 17 Given a 
particular target population, a laundry list of potential problems 
and proposed outcomes can often be identified, yet to adequately 
plan interventions priorities must be established. The promise of 
health education cannot act as a bottomless pit into which are 
placed all of society's special needs and problems for solution. 
Only those areas that are directly amenable to the interventions 
available and the resources that may be devoted to those 
interventions must be identified. Among the potential impacts of 
health education efforts, six are appropriate to the populations of 
interest. 18 

1. To increase understanding about the philosophy and science 
of individual and societal health. 

2. To increase the competencies of individuals to make 
decisions about personal behaviors that influence their 
own health. 

3. To increase the skills required by individuals to engage 
ii; behaviors that are conducive to health. 

4. To encourage the maintenance or adoption of appropriate 
health-related behaviors. 

5. To enhance the skills of individuals to maintain and 
improve the health of their families. 

6. To enhance the skills of individuals to maintain and 
improve the health of the communities in which they reside, 

Phillipp and Kolbe have identified the following specific 
priority areas directly addressable by health education 
interventions that also are directly related to excess morbidity 
and mortality among the targeted minority groups: 

• Smoking: The principal activities here should be directed 
at reducing the number of persons who start smoking and 
emphasizing the importance of stopping associated with 
particular health problems of minorities (e.g., cancer and 
high blood pressure). 



355 



• Diet and nutrition: Efforts in this area should 
concentrate principally on improving consumer choices 
given a fixed income. 

• Social support behaviors (stress, coping behaviors): 
Programs teaching social support behaviors should include 
coping with stresses associated with suicide and homicide, 
two particularly prevalent problems among minorities. 

• Exercise: Programs in this area should be designed to 
foster behaviors that can be carried through life to 
ensure fitness and that minimize problems associated with 
chronic heart disease. 

• Alcohol and drug misuse: Efforts to minimize or stop 
illicit drug abuse should also include information about 
inappropriate or non-use of medications in the treatment 
of disease. 

• Maternal and child health issues: Activities here should 
primarily be directed at enhancing early prenatal care and 
reducing maternal smoking. 

• Safety issues: Information and education about safety 
issues can range from the use of safety belts to avoiding 
occupational hazards. 

• Age at first sexual intercourse/unprotected sexual 
intercourse/number of sexual partners: Activities in this 
area should include efforts to minimize problems 
associated with sexually transmitted diseases and teenage 
pregnancy. 19 

These priority health behaviors are among the most important to be 
dealt with among minority populations, and more importantly, they 
also represent those behaviors most likely to be affected by health 
education interventions. 

C. Factors to Consider in Developing Strategies 

Because a multitude of factors can enhance or impede the 
effectiveness of a health education strategy (i.e., the creation 
and dissemination of health information and behavior change 
messages) , the development and implementation of any strategy 
necessarily entails a thorough planning process. The process 
forces the planner to consider what contributes to making the 
health message and its dissemination special for a given population 
and what may contribute to the success or failure of a health 
education program. 



356 



1. Influence of Community Leaders and Groups 

Most change takes place within the context of a social 
system. 20 Examples of social systems operating in all 
populations are the family and the community. Social systems 
can and often do facilitate or impede the diffusion of health 
information. If the nature of these social systems is not 
clearly understood by those who implement a health education 
strategy, the family and community may serve to prevent health 
information from being diffused as planned. 

On the other hand, the family and community can serve as 
tremendous resources — agents of change. Involving the family 
and community members in the strategy can strengthen the effort 
by lending credibility and visability to the activity, 
facilitating acceptance and self-determination, and creating 
greater awareness of the target community and its culture. 
Examples of people and groups at the community level who also 
might participate in a health education program include the 
following: 

• Local political leaders who have an obvious incentive to 
participate in programs designed to help the community. 

• Church leaders and visibly active church members (e.g., 
deacons, ushers ) can be effective opinion leaders, 
especially in Black, Asian Pacific, and Hispanic 
communities. Some Black churches have church nurses who 
might be especially influential as opinion leaders in the 
diffusion of health information. In many minority 
communities, churches serve as social centers as well as 
religious centers. 

• Local media and sports personalities have high visibility 
and credibility in many communities. Although some 
minority groups have little access to broadcast media, 
some communities do support minority programming. 

• School teachers traditionally are respected as influential 
members of the community, especially in Hispanic 
communities, and this role may be most prominent when a 
teacher is one of the few sources of information for 
non-English-speaking parents. 

• Alcoholism counselors can be influential leaders for 
portions of a community. In programs funded by the Indian 
Health Service, alcohol counselors often have great 
influence and credibility. 

• Club presidents can be influential opinion leaders. In 
many Spanish-speaking communities, hometown clubs consist 
of people who came or whose families came from particular 
regions outside the country. Women's clubs also are very 

357 



important in many communities, as well as fraternities and 
sororities that often have community service goals. 

• Local newspaper editors, local chapters of advocacy 

groups, and professional organizations may also provide 
effective opinion leaders at the community level. 3 

While these examples refer to any population at the community 
level, their relevance to minority populations is evident. One 
of the major benefits of involving community leaders and groups 
in the development and implementation of a minority health 
education program is that their participation helps ensure an 
accurate understanding of the target population's health 
beliefs and needs. 

2. Community Attributes 

Community traits that will influence a health education 
activity might include the relative importance placed on 
health; the cultural habits of all members or segments of the 
community; and the level of employment, which will influence 
the amount of resources available and the degree of self-esteem 
the community as a whole possesses. In addition, special 
functions may be assigned to particular members of the 
community, for example, on the basis of sex. Understanding 
these components helps ensure the most appropriate point of 
intervention as well as the nature of the message. 

3. Perceived Barriers to Taking a Health Action 

In planning a health education strategy, the planners need to 
consider whether the person or group to receive the health 
information feels susceptible to the condition or illness being 
addressed and whether they feel the condition is serious. 21 
Individual characteristics — such as fear, anxiety, a sense of 
invulnerability, self-determination, and skepticism — also 
influence behavior. 3 

In a study conducted by the National Heart, Lung, and Blood 
Institute, which examined the diffusion of information to 
culturally diverse populations, five factors were found to 
represent beliefs concerning the cause or manifestation of 
disease among some minority individuals. These included 
fatalism/God's will, naturalism, life balance, supernatural 
origin, and superstition. 3 when attempting to pursuade 
individuals to take actions to maintain or improve their health 
it is important to realize that these five factors may create a 
perceived sense of powerlessness, alienation, and inability to 
change one's destiny. While not all minority individuals 
maintain these beliefs, those who do may be less likely to seek 
out and act on messages that presume individual autonomy and 
self-control. 



358 



Another important consideration is whether the recipient of the 
information believes in the benefits to be derived from the 
recommended health behavior. 22 Belief in the effectiveness 
of a health intervention will greatly influence the likelihood 
that an individual will adopt the measure. The likelihood that 
a person will comply with a recommended health action is in 
part a function of his or her beliefs about the probable 
effectiveness of the action in reducing the threat. Perceived 
benefits are multidimensional and might include an increased 
chance for recovery, or the prevention and detection of disease 
prior to sinnptoms. Other factors such as fear of pain during 
treatment and complexity and duration of treatment serve as 
barriers for an individual considering the adoption of a health 
action. Perceived barriers can best be moderated by 
underscoring the benefits to be gained by taking the 
recommended health action. 

4. The Environment and Other Barriers to Health 

The individual and his or her behaviors and perceptions are not 
the sole source and solution to a potential or actual health 
problem. Behavioral and nonbehavioral factors contribute to 
health and disease. 1 Nonbehavioral factors that influence an 
individual's health status include environmental hazards such 
as those that exist in the home or workplace; biological 
factors such as genetic make-up, sex, and age; and inadequacies 
in the health care system. It is essential that other threats 
to health besides the behavior of the individual are recognized 
as contributing factors in any population; otherwise, it may 
appear as though individuals are being blamed for their health 
problems. Indeed, in some instances, environmental and 
technological problems may be higher priorities for and more 
apparent barriers to solving the larger health problem for the 
individual and the community. 

5. Demographic Parameters 

A variety of demographic characteristics require careful 
assessment prior to designing a health education strategy. For 
instance, the average level of education in a particular 
community, both reading skills and literacy, will dictate how 
the health message is to be shaped. In the case of preparing 
health strategies for minority populations, the level of 
acculturation — the degree and ways in which minorities adopt 
the beliefs and behaviors of the majority population — will 
influence the type of message developed. Rural versus urban 
living creates different issues related to health information 
and services. 3 

6. The Nature of the Innovation 

Examination of the attributes of the health information to be 
diffused or the health-related behavior change desired is 

359 



essential to the strategy-building process. New information, 
or innovations, can be classified according to their (a) 
relative advantage — is the innovation perceived as better than 
the existing beliefs or practices? (b) compatibility — is the 
innovation consistent with existing values, past experiences, 
and the needs of the receiver? (c) complexity — is the 
innovation perceived as difficult to do or understand? (d) 
"trialability" — can the innovation be tried on a limited basis? 
(e) observability — are the results of the innovation observable 
to the person trying the innovation and to others?3 

Problems may be encountered in any or all of these areas. For 
example, the relative advantage for a teenager to stop smoking 
may be low because the health problems associated with the 
behavior are not readily apparent. Unless the benefits of not 
smoking are firmly established, the teen will not discard the 
old behavior. Or, in the area of lowering cholesterol, it is 
not readily observable to an individual whose diet has changed 
that serum cholestrol levels are actually lowered. Observable 
results from a behavior change serve to assist in reinforcing 
the new behavior. Finally, in the area of treatment for high 
blood pressure, the initially incurred problems of 
"trialability" and complexity were overcome by developing a 
process of stepped care in treatment. Stepped care allows the 
person to progressively adopt behavior over time and thus 
integrate the changes into a daily routine. These examples 
point to the importance of carefully considering the attributes 
of the innovation as it will impact on the target population 
prior to developing a health education intervention. 

7. Channels of Communication 

Mass media, group discussions, lectures, role playing, 
modeling, community organizing, individual counseling — these 
are some of the mechanisms by which health information within 
the context of a health education program, can be conveyed to 
an individual or population of individuals. 15 An important 
element of each of these educational methods is the process of 
communication . 

Although mass media channels can be used effectively to create 
a greater awareness about a health problem, their use to create 
behavior change is generally less effective. 20, 23 Among 
culturally diverse minorities, which tend to have strong social 
networks, interpersonal channels are a preferred means for 
conveying a message to elicit behavior change. In each of the 
four minority populations, physicians are viewed as one of the 
best interpersonal channels through which to send a health 
message. 3, 24 physicians are seen as highly credible, and can 
play an important role in enhancing motivation and compliance. 
As such, physicians should be encouraged to engage in patient 
education practices that serve to improve and/or maintain the 
health of their patients. In addition, the immediate and 

360 



extended family are valued purveyors of health Information in 
an informal context. Family members can serve to encourage, 
support, and reinforce the beliefs and behaviors of individuals 
who are the target of a health education strategy. 

8. Summary of Planning Principles for Minority Populations 

A number of general guidelines can be offered as planning 
components that should be considered in the design, 
implementation, and evaluation of a health education strategy 
for a minority population. These guidelines recognize the 
value of health interventions that involve community support 
and respect differences between populations. 

- Plan a health education message that has scientific 
consensus on the validity or value of the recommendation. 

- Avoid making assumptions about the health problems , 
practices, and beliefs of any population — that is, confirm 
facts through community assessment or research. 

- Set specific, measurable goals and objectives. 

- Avoid wholesale change — fit new health messages within the 
context of existing health beliefs and practices. 

- Determine who is providing health care and health 
information within the community and its origin. 

- Assess the health needs of the target population and 
monitor community reactions to intervention efforts. 

- Involve physicians and other credible agents of change in 
efforts to convey health messages and to encourage 
behavior change. 

- Directly involve the target population in the planning, 
implementation, and evaluation phases of the health 
education strategy. In the instance of a minority health 
education strategy, (a) solicit the participation of 
bilingual and bicultural individuals of the community and 
determine preferences for the language to be used to 
convey the health message; (b) use models who are similar 
to the target group; and (c) identify the correct cultural 
context for the message or program. 

Recognize cultural diversity — for example, tribe to tribe, 
region to region, rural to urban, levels of acculturation, 
and generation gaps. 

- Use existing channels of conmiunication to facilitate the 
dissemination of information. 



361 



- Pretest health messages among members of the target 

audience for impact, comprehension, personal relevance, 
believeability, and acceptability — consider the language 
of the message, the use of illustrations, the reading 
levels of the target population, the models portrayed in 
the message, and the environment in which the message is 
portrayed. 

IV. PROGRAM ILLUSTRATIONS 

The descriptions of programs that follow are intended to provide 
the reader with a sampling of health education interventions that 
have been initiated to address the significant health problems 
facing the four specified minority populations. By no means are 
the descriptions of the programs intended to be representative or 
exemplary of all of the existing interventions of this nature. 
Rather, they serve to describe a number of approaches that have 
been employed to address minority health concerns. Although 
evaluations of the results of these projects are not currently 
available, this sampling of health education strategies is 
illustrative of programs that take into consideration those factors 
cited previously in this document as elements essential to the 
targeting of a minority population. 

A. Primary Prevention Strategies with Low Income Hispanic 
Families 

The National Coalition of Hispanic Mental Health and Human 
Services Organizations undertook this primary prevention 
project in September 1983.25 xhe purpose of the project was 
to promote awareness and action for health improvement among 
Hispanic families through risk reduction, preventive measures, 
and health education efforts. Several activities were proposed 
to accomplish the project: (1) the identification of innovative 
and promising approaches in health promotion and disease 
prevention aimed at low income Hispanic families in Head Start 
and other programs; (2) the development of health promotion 
materials for low income Hispanic families; (3) the development 
of strategies for Head Start programs to initiate or improve 
health promotion activities; and (4) the development of a 
dissemination plan to guide utilization of materials under the 
direction of the Office of Human Development 
Services/Administration on Children, Youth and Families/Head 
Start. 

This program is unique in that it will use the existing Head 
Start program mechanism, already serving a large number of 
Hispanics, as a channel through which to disseminate health 
information on such topics as stress, exercise, nutrition, 
smoking, alcohol misuse, and safety/injury prevention. The 
Head Start program will receive the following materials as a 
result of this program. 



362 



• 



A bibliography of selected consumer pamphlets, 
primarily in Spanish. Topic areas include prenatal 
care, mental health, infant care, hypertension, 
dental health, child health, safety and accident 
prevention, nutrition, and reproductive health. 

A list of curriculums that address alcohol and drug 
abuse, smoking, nutrition, exercise and fitness, 
safety and accident prevention, and stress. 

A bilingual booklet on health promotion. 

A strategy guide on how to incorporate health 
promotion activities into Head Start programs. 



B. Caide Su Corazon: Weight Reduction for Mexican Americans 

Within the State of Texas, individuals of Mexican background 
are at high risk for cardiovascular disease, in part because 
obesity is a significant health problem within the Mexican 
American population. 

To address this problem, investigators at the Baylor College of 
Medicine, National Heart and Blood Vessel Research and 
Demonstration Center in Texas are initiating a demonstration 
and evaluation project to test the effectiveness of a 
culturally adopted version of the HELP Your Heart Eating Plan 
and behavioral weight control program within a Mexican American 
community. 26 The project will assist young Mexican American 
families in developing a more active approach to family health 
and adopting dietary and physical activity patterns that 
promote (1) the reduction of risk for cardiovascular disease, 
hypertension, and diabetes; (2) the achievement of an ideal 
weight; and (3) the prevention of obesity and cardiovascular 
disease in children. 

This health education intervention will utilize social learning 
theory and social support as a basis for behavior change. Due 
to the importance of family social support within the Mexican 
American family and the familial clustering of obesity and 
other risk factors, the project will compare a family-oriented 
intervention to a more traditional individual-oriented 
treatment program. Families will be followed for one to three 
years after treatment to evaluate the long-term effectiveness 
of the intervention. 

Participants will consist of families in which one or both 
parents are 20% over ideal body weight, are between the ages of 
18 and 45, and have a child in the 3-6-year-old age range. A 
total of 180 families in two cities will participate in this 
project. Families will be randomly assigned to (1) a 
diet-booklet-only group; (2) an individual-oriented 
intervention; or (3) a family-oriented intervention. 

363 



The emphasis of the project is to assist families of Mexican 
descent in developing lasting lifestyle changes In eating and 
exercise habits, ultimately to impact on cardiovascular health, 
obestiy, hypertension, diabetes, and fitness. 

C. To Your Health - Living with Alcohol 

The Indian Nations/Tribes that participate in the Bureau of 
Indian Affairs Boarding Schools (Quechan, Cocopah, Apache, 
Navajo, Mohave, Pima, Hualapal, Supai, Hopi, and others) have 
intiated an alcohol abuse prevention project to target students 
13-18 years of age. 27 The purpose of the project is to 
reduce the incidence of drop-out, injuries, accidents, and 
arrests related to alcohol abuse and misuse among boarding 
school students. 

The project, which consists of developing and implementing a 
series of educational programs, has as its focus 
self-responsibility, limitation, and group control in relation 
to alcohol use. The series of programs will be developed and 
used within the context of Indian life, relationships, and 
social structure. This program is especially noteworthy since 
a key element of its design is the provision of an historical 
overview of Indian society and use of alcohol among Indians. 
This will include an examination of learned behavior in Indian 
society, Indian socialization practices, and the effects of 
alcohol use on Indian life, both from an individual and 
societal perspective. The overall program will emphasize the 
future of Indian life and social function in modern society. 

D. The California/Baja California Maternity Child Health Care 
Project 

Under a three year project grant from the Department of Health 
and Human Services, Division of Maternal and Child Health, the 
Health Officers Association of California (HOAC) will address 
myriad factors that affect the health status of mothers and 
Infants living in the California/Baja California boarder 
zone. 28 Based on State vital statistics, approximately 31% 
of births in 1982 in California were to women of Spanish 
origin, 89% of whom were of Mexican ethnicity. A high 
percentage of these women who are seeking and receiving 
maternity care services have low incomes. Individuals involved 
in this project are collaborating closely with the Pan American 
Health Organization, the U.S. -Mexico Border Health Association, 
and the California/Baja California Binational Health Council. 

This project will accomplish the following activities: 

• convene a series of binational meetings of health 
professionals ; 



364 



• produce a California/Baja California Health directory on 
maternal and child health programs; 

• develop an inventory of MCH health education materials; 

• design a health education campaign on perinatal care 
Issues; 

• develop perinatal health education materials, specifically 
for this population; and 

• review guidelines for the care of the low income 
Spanish-speaking pregnant women and medical treatment 
protocols for those with high risk factors. 

The California/Baja California Maternity Child Health Care 
project is unique because it focuses on developing ongoing 
communication among providers and public health officials from 
California and Baja California. The purpose of this 
communication is to foster a binational cross-cultural 
understanding of current medical nursing and related maternity 
care services practiced in both countries, including health 
education services to non-English-speaking individuals. 

E. Healthy Mothers, Healthy Babies Coalition 

Healthy Mothers, Healthy Babies is a public education effort 
carried out through a partnership among government, 
professional, and voluntary organizations and agencies. 30 a. 
coalition of members representing 66 national organizations was 
formed to: 

• provide information to promote healthful behavior among 
pregnant women and women planning pregnancy; 

• increase their understanding of health risks and the 
importance of taking personal responsibility for their 
health and the health of their infants; 

• motivate women to take action to protect their health, 
obtain regular prenatal care, and seek other counsel or 
assistance when needed. 

Low income and/or minority women are the principal targets of 
this public education effort. 

Projects undertaken by the Coalition include the development 
and dissemination of educational posters and cards to 
physicians for low income women. These cards and posters were 
printed in Spanish as well as English. Members of the 
Coalition also have developed a directory of educational 
materials, a promotional packet on breast feeding for health 
professionals, and a television production on the Coalition for 
local stations to air. 

365 



Plans for the 1985-86 calendar year include encouraging the 
establishment of State level chapters of the Coalition; 
developing a "media materials exchange network," beginning with 
New York State Health Department materials, which will be made 
available for use in 10 other States; sponsoring regional 
workshops to exchange Coalition ideas, projects, and methods; 
and developing a compendium of program ideas for motivating low 
income women to seek prenatal care. 

F. Indian Health Service Diabetes Program 

In 1979, the Indian Health Service (IHS) funded five diabetes 
model care projects as part of its Diabetes Program, to 
develop, document, and disseminate improved ways for preventing 
premature diabetes-related morbidity and mortality in Indian 
communities. 31 Training is coordinated for model site 
personnel by the IHS, and active interchange among the sites 
and other diabetes-related organizations is promoted. 
Workshops have been held to transmit the latest recommendations 
for diabetes care and education to IHS providers and to show 
providers how to implement these recommendations in the context 
of the IHS. Educational materials and approaches piloted in 
the model sites have been disseminated widely throughout the 
IHS. 

The five model projects are located in Ft. Totten, ND; 
Albuquerque, NM; Winnebago, NB; Claremore, OK; and Sacaton, 
AZ. Various types of activities that have been initiated 
within projects include emphasizing home visiting and 
teaching; providing community awareness programs and exercise 
classes; providing extensive follow-up for pregnant diabetics; 
emphasizing strong community prevention programs and producing 
culturally acceptable audio-visual teaching materials; using a 
diabetes test kitchen as a teaching tool; and developing 
programs in the areas of exercise, pregnancy, and the 
management of obese adolescent diabetics. 

V. SUMMARY 

This paper underscores the important contributions health education 
and information interventions can make to lessen the disparity in 
health status between non-minorities and minority populations in 
the United States. While the paper does not provide a definitive 
review, it does highlight some important issues to consider when 
designing and implementing a health education program for 
culturally and ethnically diverse minority populations. It is 
hoped that the preceeding discussion of issues will facilitate 
thoughtful planning and implementation of health education 
interventions for minorities in the future. 



366 



VI. OPPORTUNITIES FOR PROGRESS 
A. Information and Education 

• Several of the major health problems of minorities have 
components susceptible to intervention through educational 
efforts. Messages should be developed to target the 
special needs of the group for the given problems. 
Important issues to be addressed by such efforts include: 

• health and prenatal examinations, and positive 
maternal health habits; 

• promotion of healthy nutritional habits within the 
constraints of cultural patterns (e.g., more fiber, 
less fat, less salt); 

• deterrence of the use of tobacco products, including 
chewing tobacco and snuff; 

• deterrence of drug use, especiallly among youth; 

• deterrence of alcohol use, especially among youthful 
drivers and chronic adult alcohol abusers; 

• promotion of enhanced levels of physical activities 
for all ages; 

• enhancing awareness of hazards inherent to certain 
occupational settings; and 

• enhancing awareness of where to obtain health 
services, whether general or specialized. 

• The Department, in cooperation with the private sector, 
should assess the availability and appropriateness of 
health education materials/activities that address the 
major health problems confronting minorities: cancer, 
diabetes, violence, alcohol and drug misuse, 
cardiovascular disease, and infant mortality. Where an 
information void has been determined to exist, the 
Department and the private sector should direct a portion 
of their resources to developing culturally appropriate 
materials to meet the identified needs. Gaps in 
information exist especially in the areas of stress, 
violence, and exercise. 

• Material developed for use with a target minority 
population should be screened and tested through the use 
of focus groups or other assessment techniques with 
members of the minority population, prior to 
implementation and dissemination. Such techniques can 
also be used effectively to identify the particular needs 
of a population. 

367 



• Where materials written in languages other than English 
are needed, they should be developed from the outset of 
the program, involving professional and lay members of the 
target community in their development. 

• Innovative health education interventions should be 
developed for minorities to be used in churches, 
worksites, and schools. 

• Techniques should be explored for using pictures and words 
to convey health education messages. Such an examination 
should produce innovative ways of communicating health 
messages to culturally diverse populations. 

• A mechanism should be established to facilitate the 
exchange of health education materials appropriate for the 
identified target groups. In addition, a compendium of 
materials and a summary of selected projects should be 
developed, by disease category, which address the major 
health problems confronting minorities. 

• The Department should consider developing public 
information campaigns that target priority health problems 
facing minorities; these campaigns should be long lasting 
in nature and be evaluated for their impact. Such 
campaigns should be modeled after the successful National 
High Blood Pressure Education Program. 

• Messages developed for media use should be based on 
providing useful tools on which to act, not just 
admonitions. 

• Culturally sensitive and problem-specific materials should 
be developed, where lacking, for use in providing health 
education services to individuals in DHHS-operated 
clinical settings. 

« Providers of health education services in clinical 

settings should either come from the minority target group 
and thus possess the required cultural perspective, or be 
culturally sensitive to the needs of the target group. 

B. Access and Utilization 

• Individuals and organizations should work to ensure the 
integration of culturally sensitive health education 
programs/activities, delivered by culturally sensitive 
health care professionals, into health care services that 
target minority health problems. 

• Federal, State, and local agencies should use the 
established communication networks of organizations within 

368 



minority communities as conduits for the dissemination of 
information about health promotion, disease prevention, 
and the use of health services. 

• Because of the powerful influences of cultural factors 
developed over a period of many years in people's 
attitudes toward health behaviors, programs sponsored to 
motivate individuals to change their behaviors should be 
prepared to be sustained over time and always be 
implemented with the cooperation and participation of 
community support organizations when available. Such 
organizations can serve to reinforce the central themes of 
the program/message. 

C. Capacity Building in the Non-Federal Sector 

• The Department should consider funding risk reduction 
grants to States for programs, including those run by 
localities and private organizations, that target the 
health behavior needs of the four minority populations. 

• Many of the solutions to behaviorally associated health 
problems of minority groups have their roots in cultural 
factors which mandate carefully planned and sustained 
approaches. Health education efforts should embrace a 
three-tiered approach, including 

• development of general media-based messages on 
the target problem; 

• development of materials to be provided in 
individual counselling for persons at highest 
risk; 

• development of support networks to facilitate 
and sustain behavior change, e.g., direct 
involvement of the family, churches, employers, 
and schools. 

D. Financing Issues 

• HCFA should assess the special health and patient 
education needs of minorities, which can be provided in 
clinical settings. Once the needs have been identified, 
HCFA should implement a series of demonstration programs 
to identify the best means for reimbursing health 
education programs provided in clinical settings. 

• Federal and State governments should examine mechanisms 
for reimbursing counselling and patient education services 
provided under Medicare and Medicaid programs. 



369 



E. Health Professions Development 

• A series of training seminars on health education and 
health promotion techniques/methods for minority 
populations, directed at practitioners within the National 
Health Service Corps and those working under HHS grants 
should be sponsored. The seminars should utilize the 
existing resources and expertise of the Department of 
Health and Human Services. 

• The public and private sectors should examine ways to 
increase the minority representation in health education, 
communications, and other health professions. 

F. Leadership, Work with Other Sectors 

• The Department should convene a meeting with leading 
minority organizations to chart a strategy for diffusing 
health information among minority groups. The involvement 
of major voluntary organizations in this activity is 
recommended. By involving a broad array of agencies and 
individuals, the Department could mobilize and strengthen 
private and public efforts to address minority health 
information needs. 

G. Research Issues 

• Impact and outcome evaluations of minority health 
education interventions should be sponsored to help plan 
or modify intervention and to justify the allocation of 
resources to such projects. 

• Research studies among the four minority populations 
should be sponsored to identify more accurately existing 
health beliefs and practices and to determine what are 
their sources of health information. 

• Research needs to be conducted that will elucidate the specific 
characteristics of minority populations that may impede or 
facilitate the diffusion of health information. 

• Researchers should examine different ways of approaching 
minority populations to affect behavior change. For example: 

• How do we motivate young Black males to change their 
behavior in the area of violence? 

• If there are male and female differentials in behavior 
patterns, why do they exist and what is the nature of those 
differences? 

• What is the nature of minority populations' dietary 
behaviors? 

370 



H. Data Issues 



The Department should develop a data collection model for 
application at the local level, so that communities can identify 
their own needs in order to develop interventions at the local 
level. 

The Department should work with the non-federal sector to 
develop a data base on health needs, health beliefs and 
practices, and sources of health information among minority 
populations. 



371 



VII . REFERENCES 

1. U.S. Department of Health and Human Services, Healthy People: 
The Surgeon General's Report on Health Promotion and Disease 
Prevention^ DHEW (PHS) Pub. No. 79-55071 Washington, D.C.: 
U.S. Government Printing Office, 1979. 

2. Personal Communication, National Center for Health Statistics, 
October, 1984. 

3. U.S. Department of Health and Human Services. Development of 
Diffusion Strategies Among Culturally Diverse Populations , NIH 
Pub. No. 84-2697, National Institutes of Health, National 
Heart, Lung, and Blood Institute: 1984. 

4. U.S. Department of Health and Human Services. Health Education 
and the Black Community . Centers for Disease Control: 
December 1980. 

5. Juarez and Associates, Inc. Healthy Mothers Market Research: 
How to Reach Black and Mexican American Women . Office of the 
Assistant Secretary for Health, Public Health Service, U.S. 
Department of Health and Human Services, Washington, D.C.: 
September, 1982. 

6. U.S. Department of Health and Human Services. Proceedings of 
the Conference on Communicating with Mexican Americans: Por Su 
Buena Salud. National Institutes of Health: June 1981. 

7. Ramirez, A.G., and Cousins, J.C. Hispanic Women's Health 
Issues: Understanding a Mosaic Population. Paper presented at 
the American Public Health Association Annual Meeting, November 
1983. 

8. Ramirez, A.G., Gombeski, W.R., Kantz, J. A., Farge, E.J., Moore, 
T.J., and Weaver, F.J. Communicating Health Information to 
Urban Mexican Americans: Sources of Health Information. 
Health Education Quarterly 9(4): 293-309, Winter 1982. 

9. da Silva, G.C. Awareness of Hispanic Cultural Issues in the 
Health Care Setting. Children's Health Care 13(1): 4-10, 
Summer 1984. 

10. Warneke, R.B., Intervention in Black Populations. Cancer Among 
Black Populations . New York, N.Y.: Alan R. Liss, Inc., 1981. 

11. U.S. Department of Health and Human Services, Indian Health 
Service Chart Book Series . Public Health Service, Health 
Resources and Services Administration, Washington, D.C.: June, 
1984. 



372 



12. Personal Communication, U.S. Department of Health and Human 
Services, Indian Health Service, January 5, 1985. 

13. Personal Communication, San Francisco General Hospital, Robert 
N. Ross Patient Education Resource Center, December 19, 1984. 

14. Hochbaum, G.M. Health Behavior. Belmont, CA: Wadsworth 
Publishing Company, Inc., 1970. 

15. Green, L.W. , Kreuter, M.W. , Deeds, S.G., Partridge, K.B. , 
Health Education Planning: A Diagnostic Approach , Palo Alto, 
CA, Mayfield Publishing Company: 1980. 

16. Green, L.W. , Health Information and Health Education: There's a 
Big Difference Between Them. Bulletin of the American Society 
for Information and Science 4(4): 15-16, 1978. 

17. Kolbe, L., and Iverson, D. Comprehensive School Health 
Education Programs. Miller, Matarazzo, Weiss, Herd, and Weiss 
(Eds.), Behavioral Health: A Handbook of Health Enhancement and 
Disease Prevention , New York, N.Y.: John Wiley & Sons, 1984. 

18. Kolbe, L., Predicting the Impact of School-based Health 
Promotion: Some Variables in the Formula. Presentation to the 
NIH Health Promotion Subcoimnittee, National Institutes of 
Health, Bethesda, Maryland, February, 1985. 

19. Phillipp, A., and Kolbe, L. Changes in the Prevalence of 
Priority Adolescent Health Risk Behaviors . In press. 

20. Rogers, E.M. and Shoemaker, F.F., Communication of Innovations; 
A Cross-Cultural Approach , New York, N.Y.: MacMillan Publishing 
Co., Inc., 1971. 

21. Rosenstock, I.M. , Historical Origins of the Health Belief 
Model, Health Education Monographs 2, 328-35, 1974. 

22. Kirscht, J. P., The Health Belief Model and Illness Behavior, 
Health Education Monographs 387-408, 1974. 



373 



23. McQuail, D. and Wlndahl, S. Communication Models for the Study 
of Mass Communication , New York, N.Y.: Longman, Inc., 1981. 

24. U.S. Department of Health and Human Services. Strategies for 
Promoting Health for Specific Populations . DHHS(PHS) Pub. No. 
81-50169, Office of Disease Prevention and Health Promotion: 
1981. 

25. Personal Communication, National Coalition of Hispanic Mental 
Health and Human Services Organization, December, 1984. 

26. Personal Communication, Department of Medicine, Baylor College 
of Medicine, January, 1985. 

27. Personal Communication, Indian Health Service, January, 1985. 

28. Personal Conmiunication, Health Officers Association of 
California, January, 1985. 

29. Personal Communication, U.S. Department of Health and Human 
Services, Public Health Service, Region V, January, 1985. 

30. Personal Communication, U.S. Department of Health and Human 
Services, Public Health Service, Office of the Assistant 
Secretary for Health, January, 1985. 

31. Personal Communication, U.S. Department of Health and Human 
Services, Indian Health Service, February, 1985. 



374 



Minority and Other Health 
Professionals Serving 
Minority Communities 

Report of the 

Working Group on Health Professionals 



MINORITY AND OTHER HEALTH PROFESSIONALS SERVING MINORITY COMMUNITIES 
Report of the Working Group on Health Professionals 

Contents 

Acknowledgements 378 

Introduction 379 

Part I. Summary and Conclusions 382 

Part II. Selected Variables Influencing Health Professionals: 

Numbers, Types, and Distribution 394 

A. Minority Populations and Communities and their 
Health Professional Resources 

1. Black 

2. Hispanic 

3. Asian/Pacific Islander 

4. American Indian 

B. Minority Health Professionals 

1. Distribution 

2. Development 

3. Practice 

Part III. Why do the Differences Exist and How do They 

Contribute to the Health Status Disparities? . . . 484 

Appendices 

I. Counties with 20 percent of Population in Any 

One Minority Group 495 

II. Minority Health Professions School Graduates .... 509 

III. The Treatment Practices of Black Physicians: Summary . . 541 

IV. Letter to External Community 545 



377 



ACKNOWLEDGEMENTS 



This report represents a coordinated effort under the guidance of the Bureau 
of Health Professions, Health Resources and Services Administration, Public 
Health Service. We, the working group on Health Professionals, gratefully 
acknowledge the following individuals for their contributions: 

William A. Darity, Ph.D., Professor of Public Health and Dean of the 
School of Health Sciences, University of Massachusetts at Amherst for his 
research, editorial and technical assistance, 

Everett R. Rhoades, M.D., Director, Indian Health Service, for his counsel 
and technical support, 

Howard V. Stambler, Director, Office of Data Analysis and Management, and 
his staff including, Roger B. Cole, Chief, Information Systems Branch, 
Leonard A. Drabek, Chief, Technical Analysis and Coordination Branch, and 
Ernell Spratley, General Statistician, for the development of information 
on the U.S. and minority populations and the U.S. health professionals, 
largely through their Area Resource File, 

Vivian Chen, Program Analyst, Division of Medicine, Remy Aronoff, Program 
Analyst, Analysis and Evaluation Branch, Division of Disadvantaged 
Assistance and Blake C. Crawford, Writer-Editor, Division of Associated 
and Dental Health Professions, all of the Bureau of Health Professions for 
compiling data, graphs and tables, and developing narratives to present 
this information on minority health professionals, and communities. 

Additional appreciation is extended to Marcella Murphy, Medical Services 
Assistant, Commissioned Personnel Operations Division, Public Health 
Service, Leonora Surosky, Secretary, Kathy Owens, Secretary, Stacey L. 
Williams, Clerk-Typist and Deborah A. Hunter, Clerk-Typist, Division of 
Disadvantaged Assistance, Donna Breslyn, Clerk-Typist, Division of 
Medicine and Pamela Dobson, Clerk-Typist, Division of Associated and 
Dental Health Professions for their clerical and secretarial support in 
preparing portions of this document. Special appreciation to 
Susan Eddins, Secretary, Office of the Director, Bureau of Health 
Professions, for her assistance in all support phases throughout the 
development of this activity and the preparation of the final report. 



William A. Robinson, M.D., M.P.H. 
Clay E. Simpson, Jr., Ph.D. 
Frank A. Hamilton, M.D., M.P.H. 



378 



INTRODUCTION 



BACKGROUND 

During the process of developing this broad study to examine the persistent 
differences in health status between the nonminority and minority U.S. 
populations, it became apparent that several cross-cutting factors were 
probably contributing to these differences. Some of these factors were the 
relative socio-economic status of the population groups; their dietary habits 
and nutrition status; the availability and accessibility of health facilities 
and personnel; occupational and environmental conditions; and others. At the 
direction of the Task Force Coordinating Committee, working groups and 
subcommittees were established to attempt to analyze the influences of these 
individual cross-cutting factors. 

This report was developed by the working group on Health Professionals, whose 
charge was to examine the importance of the availability of health 
professionals as a factor influencing the disparities in health status between 
nonminority and minority communities. Keeping in mind that other working 
groups would be addressing such issues as access to health resources, and the 
financing of health care as other important factors, this report was to 
address the following questions: 

• What are the variables concerning health professionals that create, 
enhance or foster differences between the nonminority and minority 
populations? 

• How do these differences subsequently contribute to the evidenced 
disparity in health status? 

• Why do these differences exist? 

In further consideration, the working group was also asked to respond to these 
questions: 

• "What are the patterns of (a) minority health professional 
development, distribution, and practice, (b) health professionals 
serving minority communities, and (c) minority health professionals 
serving minority communities?" 

• "What do health professionals expect of their clientele, and what does 
the client expect of the professional? Can expectations be meshed and 
if so how?" 

In determining the availability of health professional resources, the working 
group chose to attempt to analyze the following variables: 

(a) the numbers and types of health professionals appropriate to address 
the major health problems under study, and 

(b) the geographic distribution of these individuals in comparison to the 
location of minority population groups. 



379 



In an effort to make the analysis of "numbers, types and distribution" most 
meaningful for the diseases and conditions under study (e.g. cardiovascular 
disease, cancer, etc.), an initial attempt was made to develop a list of 
health professionals involved in providing health care for a sample patient 
encountering the health system with a single, reasonably common sample 
condition. Using a non-fatal traffic accident (violence) as an example, the 
list might include, among others: 

Emergency Medical Technicians 

Emergency Room Specialists and Nurses 

Medical Records Staff 

Clinical Laboratory and Blood Banking Personnel 

Radiologists and Radiographers 

Anesthesiology Specialists 

Operating Room Nurses, Technicians, et al. 

General and Specialty Surgeons 

Intensive Care Nurses and Staff Nurses 

Medical Social Workers 

Physical Medicine and Rehabilitation Specialists 

Occupational and/or Physical Therapists 

By extrapolating to include each of the health problems under study, in a 
variety of settings, it seemed apparent that virtually all of the different 
types of professionals would eventually have to be considered if a 
comprehensive study were to be undertaken. Such a study is planned. 

The multiplicity of personnel engaged in providing health care, while critical 
to the overall well-being of the patient also makes it particularly difficult 
to demonstrate that the role of any one (or few) is the keystone to improved 
health status. 

If the task had been to review the availability of health professionals in 
general, it would be comprehensive and require review of large amounts of 
data. Fortunately, systems providing such data on a national, regional, and 
often state-wide basis have already accomplished basic data collection and 
analysis functions. [One of the most comprehensive is the Bureau of Health 
Professions' Biennial Report to Congress on the Status of Health Professions 
Personnel. ] 

The assigned task, however, was to analyze the availability of health 
professionals to the minority communities , thereby significantly complicating 
the review process. The first obstacle came in attempting to define a 
minority community in such a way that the subject minority group (Black, 
Hispanic, Asian/Pacific Islander or American Indian) could be analyzed within 
specific geographic boundaries for which there were also data on health 
professional resources. The system identified to facilitate the definition 
and analysis processes was the Bureau of Health Professions' Area Resource 
File (ARF). 



380 



The second obstacle to the task was encountered in trying to collect data on 
minority health professionals, especially data that could be analyzed in the 
context of the previously defined minority communities. Although statistics 
on the development of minority health professionals are reasonably available, 
information on their distribution and practice is poorly accessible, if at all 
available at the "community" level. 

The third, and perhaps most dominant constraint on this report was the time 
alloted to accomplish the work. As a result, rather than this effort 
comprising a definitive study, it constitutes the first phase or beginning of 
what will, hopefully, evolve into a more detailed and complete examination of 
the health professional resources available within minority communities. 

THE REPORT 

Because of the interrelationship of the two sets of questions posed to the 
working group, this report has been structured to address them concurrently. 
The first section of the report presents the conclusions drawn, while 
summarizing the highlights of the report. Needs for further intervention 
and/or monitoring are also described, within the context of "Recommendations." 

The second, and most quantitative portion of the report, presents descriptive 
information on the four minority populations (and "communities"), and the 
health professionals who serve them (both nonminority and minority). It 
discusses who they are and where they are, emphasizing differences between the 
minority and nonminority populations and their resources. 

The data included in this report were derived from numerous sources. Although 
there is some consistency for cross referencing information, most data sources 
generated their own terms, or groupings of minorities, or other indices. 

The third portion of the report attempts to focus on the more subjective parts 
of the task, including trying to answer the "how" and "why" questions, and 
discussing health professional/provider - client expectations. 



381 



PART I. - SUMMARY AND CONCLUSIONS 

This report represents the completion of the first phase of a study to examine 
the availability of health professionals to minority communities. 

The effort began with the examination of the minority groups themselves, first 
at the national, then state, and finally county level to attempt to define 
boundaries for a "minority community." The progression of the review from the 
larger to the smaller geographic units was designed to identify the largest 
unit which might reasonably be called a "community," for which health 
professional resource information was available or could be quickly generated. 

Because of the differences in the sizes and the patterns of distribution of 
the minority groups, no single standard for comparing all of them at the 
state/county level could be agreed upon. As a result, the working group 
arbitrarily selected to examine data for U.S. counties having a population of 
greater than 20 percent for any of the minority groups. [A lower density of 
five or ten percent was used for some Asian analyses.] The review focused on 
analyzing the attributes of those counties taken collectively within a state 
in comparison to the rest of the counties within that state. The cluster of 
counties was to serve as a proxy for a "community." 

The analysis of state/county level data was directed first towards the 
demographics of the respective minority groups, then towards the issue of 
health professional resource availability. Although the socio-economic and 
related characteristics of the target (key) counties were generally consistent 
with the national picture of these minorities (e.g. higher infant mortality 
and poverty rates), these features could not be ascribed specifically to the 
minority groups in question. Further, there was no mechanism for examining 
differential characteristics of minority sub-groups, particularly among 
Hispanics and Asian/Pacific Islanders. These same problems carried over into 
the review of resource availability. 

The methodology did not allow for full examination of variances due to urban 
versus rural environment; the specific influences of the presence of other 
minority populations; features of the general population; or several other 
potentially significant factors. 

Much of the discussion of the availability of health professionals focused on 
physicians. While inappropriate as a sole measure of resource availability, 
physicians were felt to be appropriate as a partial barometer, especially 
since the methodology was still evolving, and trends for other professions, 
where available, followed similar paths. Generally the data suggest: 

• Many high density minority counties have substantially lower numbers 
and percentages of health professionals than their low-minority 
counterparts. 

• The Indian Health Service plays a critical role in providing needed 
resources in many American Indian communities. 



382 



• Where high density minority counties have numbers and percentages of 
health professionals which equal or exceed the other counties, it is 
generally found to correlate with the location in those counties of 
significant health professions training resources. The numbers of 
health professional faculty usually overstate the availability of 
practitioners for patient care. 

• More specific data and analyses are required to answer specific 
questions on specific communities and sub-groups, however some more 
accurate generalizations can probably be achieved through 
re-structuring of existing county-level data analyses. 

Data on minority health professionals are quite mixed in availability, 
specificity and reliability. Student data are generally much better than that 
on practicing professionals, but both need to be more closely analyzed for 
relevance to the minority sub-groups and communities. With some notable 
exceptions among sub-groups of Hispanics and Asian/Pacific Islanders, 
minorities are substantially underrepresented among students and practitioners 
of virtually all major health and allied health professions disciplines. As a 
consequence they are poorly available as a major resource to assist their 
respective communities. 

The same patterns of paucity/scarcity persist among minority health 
researchers and science faculty in health educational and training 
institutions. This may be especially crucial and have negative consequences 
for rate and degree of progress being made in improving the health status of 
these minorities and their communities. 

Although definitive evidence is not available, existing studies of 
interpersonal and sociologic behavior suggest that the availability of health 
professionals who are from the same cultural background as their patients, may 
provide for a greater awareness of and sensitivity to the health-influencing 
factors which determine a positive health outcome. More study is required to 
evaluate the importance of this area to the health status of individual 
minority sub-groups. 

Some minority-specific summary comments are provided below. 

Blacks 

• The Black population, although perhaps not as diverse as some other 
minority populations, does encompass several sub-populations. These vary 
not only by socio-economic status, urban versus rural residence and 
educational attainment, as customary documentation suggests. They also 
vary by factors such as the homeland of their parents/grandparents (the 
U.S., Africa, the Caribbean), religion, family customs, etc. 

• In the 23 states and the District of Columbia, which provided the focus 
for this report, the heavily Black counties have larger poverty-level 
populations and larger number of Aid to Families with Dependent Children 
(AFDC) recipients per 100,000 population. The percentage of AFDC 
recipients was generally higher in urban Black counties. 



383 



Black Americans have higher infant mortality than the overall population. 
In virtually all states, both in the key counties and elsewhere, Blacks 
had higher infant mortality rates than nonminorities . The single 
exception was for those Blacks living outside the key counties within the 
State of California, where Blacks had a marginally better infant mortality 
rate than nonminorities. 

General hospital utilization figures show that inpatient days per 1000 
population were higher in the key counties of 22 of 23 states. 

More than half of the Black population of the target area lived in urban 
counties, which had professional to population figures slightly or 
substantially below national and national urban rates (with lower rates 
for counties with higher Black concentrations). Rural counties with Black 
populations of 20 percent or more had professional-to-population ratios 
(for all professionals) well below the figures for all rural counties and 
far below national averages. 

Black-physician-to-Black-population ratios were significantly lower than 
the overall physician-to-population ratios for the key counties in all 23 
states, although 19 of these had higher overall physician-to-population 
ratios than the remaining counties. Only in the District of Columbia and 
the key California counties were there as many as 100 Black physicians per 
100,000 Black population. All other states were far behind. The picture 
appears similar for other health professions. Aggregate national data on 
dentists. Registered Nurses (R.N.), optometrists, and pharmacists also 
show consistently low Black-prof essional-to-Black-population ratios, well 
below professional-to-population ratios for the nonminority population. 

The proportions of Black health professionals serving Black populations 
are not likely to change appreciably in the near future. In virtually 
none of the states considered does the percentage of Black graduates of 
medical, dental, and pharmacy schools (the three disciplines examined) 
approach the percentage of Blacks in the population. Thus, even if the 
numbers of Black graduates continue to rise, it is not likely that they 
will significantly alter Black professional-to-population ratios in the 
near future. 

The relative paucity of Black and several other minority health 
professionals has been documented, with implications not only for their 
availability to provide care to their communities, but to participate as 
faculty in schools which train those professionals; and to engage as 
researchers and scientists in studying the problems which affect their 
communities. 

All of the above suggest, as it did for other minority groups, the need 
for more specific, targeted attention to the individual Black communities, 
to not only address the availability of health professions resources, but 
also the general adequacy of health services. 



384 



J 



0) 

re 
*-> 

>- 
09 



c 
o 



3 
Q. 
O 

Q. 

u 
_re 

m 

o 

00 
O) 




D. 
O 

a. 



o 

E 



O 
CM 



c 
o 



3 

a. 
o 
a. 



00 

O 
CM 

C 



385 



Hispanlcs 

The Hispanic population is generally concentrated in only nine out of 
fifty states. 

A critical factor which becomes problematic when reviewing the data on the 
"Hispanic communities" is that they are actually a group of communities 
brought together for analysis because of a common ethnicity and language 
base. There are important demographic, life-style and other differences 
among these communities, varying with origins in Mexico, Puerto Rico, 
Cuba, Europe and Central and South America. Each sub-population 
eventually must be examined as independently as possible to develop a true 
picture of availability and accessibility of health professionals. 

Further, because "Hispanic" denotes ethnicity rather than racial 
identification, collection of data relevant to these populations has 
lagged significantly behind some other groups. Cultural and language 
variations have either been difficult to define or considered not of 
sufficient importance for inclusion in many data collection efforts. 

Another factor found in analyzing data for the Hispanic community and 
health professionals may be the presence of significant numbers of 
individuals who are not citizens. In the field of medicine, for example, 
of the 3,655 Hispanic physicians in residency training in September 1983, 
more than 1,700 were graduates of foreign medical schools. The 
implications of the presence and contributions of these individuals, 
particularly those whose non-citizen status introduces other variables, 
should be the subject of further study. 

The availability of health professionals to Hispanic communities seems to 
present a mixed picture. The widespread variation in residence patterns 
ranges from dense urban populations in some states (California, Florida, 
Illinois and New York are some examples) to more sparsely populated rural 
areas of others (New Mexico, Texas and Colorado are other examples). 
Further, there are mixed patterns even within these states, which 
complicate the picture even more. Some areas seem to have large numbers 
of health professionals available, and in a few counties even large 
numbers of Hispanic health professionals. However it has not been shown 
that these health providers are geographically spread so as to be made 
available to all those Hispanics in those counties. More detailed review 
of specific county level data, followed by analysis of selected sub-county 
level data is necessary to provide clarity to the circumstance of the 
individual sub-populations of Hispanics. 

Mexican-American and Puerto Rican Hispanics appear to continue to be 
under-represented in the education and practice of various health 
professions but in many professions the picture is clouded by the possible 
over-representation of other Hispanics. More definitive data analyses are 
required. 

Other national, state and education data need to reflect the existence of 
the sub-populations which comprise the Hispanic minority, and the degree 
to which they have available resources to provide improved health and to 
pursue health professions careers. 



386 



I 

1 



0) 
CO 

> 

vi 

D 

0} 



c 
o 



3 

a 
o 

Q. 



C 
Q. 



O 
00 
Oi 




c 
o 



D. 
O 
O. 
U 

c 



o 



a 
o 
a 





O 

CN 

c 


.0 


CD 


X 


*-• 






o 


to 

a; 


CO 


^ 



c 

O 



387 



Asian/Pacific Islanders 

Demographically, Asian/Pacific Islanders tend to live in dense populations 
in a few states (Hawaii, California and New York), but in much smaller 
concentrations in several other states. In fact these three states are 
the only ones containing counties in which Asian/Pacific Islanders 
constitute more than 5 percent of the population. As a consequence, 
analyses of data at the county level regarding health status, numbers and 
types of health personnel, etc. were made ineffectual by the small 
population figures. 

Hawaii is by far the largest single Asian/Pacific Islander community in 
the U.S. However, because of its size and its political definition as a 
State, it was deemed inappropriate to include that State for bases of 
comparison with other Asian/Pacific Islander conraiunities. 

At least three critical factors complicate the examination of data on 
Asian/Pacific Islander populations and communities. One is degree of 
heterogeneity which the term "Asian" masks. The residents of these 
communities have origins in a large number of countries and cultures from 
throughout the continent of Asia, and include many island nations of the 
Pacific. Many are well-known and live in larger communities (e.g. 
Chinese, Asian Indians, Philippines, etc.) while others are lesser-known 
or more recent immigrants living in smaller communities (e.g. Laotians, 
Samoans). They encompass a broad spectrum by socio-economic status and 
other demographic variables. Most U.S. data collection efforts do not 
provide a mechanism for analyzing these sub-populations. 

A second critical factor that is frequently encountered among the 
Asian/Pacific Islander communities is the geographic proximity to other 
minority communities such that the latter, and usually larger, minority 
group dominates the limited focus on "minority concerns." County level 
statistics are generally not discrete enough to provide meaningful 
information on Asian/Pacific Islander populations. Sub-county level data 
will have to be examined to obtain a realistic picture. This is 
particularly so when attempting to look for differences between Asian 
sub-groups. It is also important that eventually Asian/Pacific Islander 
sub-groups be compared to other Asian/Pacific Islander communities and not 
solely to the nonminority or large "other minority" communities. 

A third factor which must be considered in the analysis of data on 
Asian/Pacific Islander health professionals and their communities, is the 
presence of significant numbers of individuals who are not citizens. In 
the field of medicine, for example, of the 5,632 Asian/Pacific Islander 
physicians in residency training in September 1983, more than 4,000 were 
graduates of foreign medical schools. The implications of the presence 
and contributions of those individuals, whose non-citizen status 
introduces other variables, should be the subject of further study. 

In broader analyses, Asian/Pacific Islanders as a group appear to be 
disproportionately over-represented both in the education/ training and 
practice of the health professions. Only to a limited degree have data 
been available to demonstrate which sub-populations contribute to that 
over-representation by profession. 



388 



The data which have been analyzed are not sufficient to provide definitive 
statements on the availability of health professionals to Asian/Pacific 
Islander communities, especially to sub-populations of the "Asian/Pacific 
Islander" minority group. 



389 



1980 Asian Population in the State of California, by County 




Less than 5% 
5% and over 



390 




391 



American Indians 

American Indians represent the smallest of the four minority groups with 
total numbers, by 1980 Census count, of approximately 1.4 million. This 
represents a 70 percent increase over the 1970 Census, which undoubtedly 
is due to increased numbers of people declaring this identification, 
rather than the result of changes in vital events, i.e. births and deaths. 

Counties having greater than 20% Indian population are spread over 10 
States. In part this is a reflection of the location of reservations; 
otherwise it reflects strong tendency to live as community in single 
counties rather than dispersing. 

Generally, in addition to being poorer, Indians suffer from increased 
health problems, such as diabetes and cirrhosis. They also have generally 
lived in communities where health professionals and other health resources 
were deficient. Fortunately the Indian Health Service has compensated for 
some of these deficiencies, in selected communities, but real problems 
remain. 

Despite relatively small numbers there is a great deal of heterogeneity in 
the American Indian population, primarily due to the existence of more 
than 300 tribes. Customs and cultures vary widely. The nature and 
formality of the relationships between these tribes and communities, and 
local. State and Federal governments also show wide variation. To address 
the concerns of the individual Indian communities will require information 
on the specific make-up of the residents of that coimnunity and inevitably, 
sub-county and tribal data will have to be obtained and analyzed. This 
further suggests it might be appropriate to compare certain Indian 
communities to other Indian communities, and not solely to the nonminority 
or other larger minority populations. 

Another major factor which must be considered in any review of Indian 
communities is the varying presence and role of the Indian Health 
Service. More time and specific data are required to complete the review 
of the role of this important resource. 



392 



0) 

*-> 

SI 
0) 



c 
.2 
"+^ 

JO 

3 

a 
o 

o. 

c 
.2 
■5 

c 

c 

(O 

_o 

0) 

E 
< 

o 

00 




3 

a 
o 
a 



c 



E 
< 



o 



a 

o 

Q. 



■D 

C 



o 

CN 



c 

O 



393 



PART II. - SELECTED VARIABLES INFLUENCING HEALTH PROFESSIONALS; NUMBERS, 
TYPES, AND DISTRIBUTION 

PART II. A. - Minority Populations and Communities 

The principal reason for taking a look at minority communities in the context 
of this working group's charge, was to define the boundaries for comparisons 
of the relative availability of health professional resources between minority 
and nonminority communities. Other reasons are that the demographic profile 
of the population groups and the environments in which they live should be key 
factors considered when analyzing the need for and use of various health 
professionals. 

Some of the differences between and among the minority populations, are being 
discussed in other working group reports. Nonetheless, some re-statement of 
the differences, and the demographics, is presented to facilitate the 
discussion of this topic. 

To define "community" for each of the four minority populations the first 
activity was to identify an appropriate sub-population of the total U.S. 
population of each of those groups. Listings of the States were generated, 
based on 1980 Census data, through the Bureau of Health Professions' Area 
Resource File (ARF). A review of the data derived confirmed that the four 
groups were broadly spread throughout the 50 States, each with its own pattern 
of dense and sparse concentrations, and each different from the nonminority 
population. 

Table 1 displays the Distribution of the U.S. Population by State and by 
Race/Ethnic Category, while Table 2 displays the Distribution of the State 
Populations by Race/Ethnic Category. Tables 3A-D display Rankings of Twelve 
States Having Largest Population of Individual Minority Group, by Number and 
Percent. 

Because the statistics showed that only the State of Hawaii (and the District 
of Columbia) among the States had a significant enough minority population to 
have that state be possibly deemed a "minority community," a further analysis 
was conducted of county-level population data. 

Through further use of the ARF System, a listing was generated of all U.S. , 
counties having a population of at least twenty (20) percent in any one of the 
four minority groups. The 20 percent level was selected as having enough of a 
critical mass to meet a general concensus definition of "community," with the 
exception of Asian/Pacific Islanders, as noted below. See Appendix II. A. 

Table 4 displays a listing of the 50 States showing the number of counties 
within each, with a population of at least 20 percent in one of the four 
minority groups. 



394 



r- 






»ii\«««^oeooe»ea" 



i^eBO«o*>oo^oa^eea^aoeooeec 






i(r>co(NJ^r^knoiiOro%r^tAevors.oro<«r»eitf*-rMeo^r^CNjoo^iOK) 









»ee^*-otno>^e^^eoo^ ■ 



>OOO^OOCM^<OOO^O^^OOOOrO^^'l< 



^o^r^Jr^JlOr>•^or«-po•J•^<^^Kttf^r^•9''OPOf09•l^l^«o^(^J^^o^'4'f^eo<^Jec^J<«u^«o^r>•or^f^la^oo^r^(^l»o^ 






mK)r»r\irNj(Mr^<iffOtoiriKi9>r^etocMroo>o^<CfO*-in^o^e*tf>NCNj 
csj ^iTt^fo CM ^ e porg irt^*rr^ — — iocm r^— cm 





Zl 








■< 




m 


H- 


o 


t- 


c 


Q. 


«c 


z 




»- 








irt 




Z 




z 


G 




>- 




»- 




n 


c 


t- 






t~ 


■< 




« 


0£ 






z 


IL 


r 




o 


r 


0. 






■< 


o 
a. 








^<ro^in^ooo^e^e^oo*-oaooac\iCNfooN^oao>ocM^^ — < 



- o o <^ e NO 



a. 
o 

0. 

u 

z 



H 



^ff««r»rrj — — •a 

m — CSJ ^ — 

osou^oo — r^i0CJ<iO9*^K)rs. — (Mp«.to«oeo«o**NtocMoin^^coO'CJO^^coeo*-r«-^^ino«o— •^«f<TO"-omo< — 
fO^^^<Aeo^^iAirtoe<o^eoettf-oK>osroK)«-^oeefOeo«sro^eeroefOocst<ooeK)^eee 

(r>rooo9>moin*r^o^iArs.^rx.a^«0'«o^(M*-flO<M^irt— meo«AiOsr9^*~ oO"*- — fMo^no— ff»fjoo — ^oOin»r— ff»o 

00 ^ — ^- r^ ■- o •* o* NT <*■ so ^ r* — sj- <sj in K) u^cMO<>m«o— -^m pjcmo— ps. o k> st rj •*■ pj^ oo^to 

*ff» fOflO— -tNi ^m*T ^i'-T — C4CSJ OTJ*- to m o* ^lo ocs* o a» r^-r^ o^ — 

^ •- ^-^ w~ ^ ^ CM— w- wm ^ *. 



(L ^O*- — ^^ — ^0-^fSIOO*CNJ^— ^ — ^O—CNJ^rjOCMOOOOPOOP-fNJOm^' 



i-i UJ z 



It 3 
Of 



ocjr^oo^r^a^>«9>fOOkcocoaiAoinc^<^(^^tn«o<Mneo^9^r^i0^o(M(Mt«K)«o«0(SJ^«o>f^roaoKir»<tf-«-9>m^ 



-ocNJinooNrsjorsio 



cvi<»»««oirtK>ocjcsjiftinK>iAinrvr*io^**or*.cj»oooirifsjrj^oK»^K»r*.>rooof^fOoO'Oootrtooo**r^pJino<»«)^ 
org9«rsi9>>or\ioersiflOtrio«o>oootoarjK>(r'<'^c\jo<«^av(Nie(neO'«>«irtro--eK)>d->j-«oeo^o^a'Oa«<oo^ 
lAro— o<7> — — — «o^«ro^^^r->><o— eou^c>jsoaorocjinooo(Njroo>rotoevor>.K^'«eoer<<>(^r<Siioror^%oo-.r<o>0 



a« C\i CM — OOCNJCM 



<Mft CM CM »0 CM •- iO tn r*. KJ — <»■ »- 



«y^r (MMCMO CJ Kl — ^ 



oCM«o^*~tA«or^^<eo — K)CJ<^CM(^<^CMtde<fO(^o^^irt<r«(n<-'meomo9ir^^>D>TfMeoorooO'^'0>o(M^iO — 

Or«. — (MO^TCM'OCMfMtO^^J■^O^CM^■*lOl'^iOl^OOO— — '0*OK»«,rCJmr»*'*CMr^rO"-fM^»0»OOCM^CM»0€0«00<M 



.^'—^-O— N<tf'«-< 



■j^r^CMo^ — — i 



-OCM*«OOCM*-OCMO 



inoo — mmcsi<<r^<Of0^in^iri«o^riO<^f^^c»irtr«-Maco^ooiriK)eio<^CM«Arveointn>raooo — r<»ii0ao^<tf-r^r^ 

e>«)iri'"io^<orN.KifO{Mo<F'K>— CMO^^r^o^^^.^o^^^*>.^ooo<^CJC7*^f^Jff*r**^^■'0^off»l/lC\J^ocMc^rOln— if>*T'«oin 

J>-«| ■o«oooCMsr^^»f»fOioiO'^^ff»incM«)^r*W'*ff'oo*"0"«*o«0''''*«0«oor««.rN..ocM'-«o*"cor«.^ — o*T«j^>4>r*ift 

inK>— •o*or^<7*r^'^«0'«»'i^»o<oo»OK>oifi»»->fl^.<Mif»^>a'*<T*oo^csioO'*CMP«-ir(»OK>r«-— o — <^^^*oeM«^ift*> 

<f0<^^M*Ai0O0^r0i^tOtO<^CM0*''<O'O^CM — ''^^'^(M^'>O'4^<M>O^tAaOirt(^rsjK>W3%^(MO^O^(M^**<fO<Xe«0 

in«0'ifr*>cj<iA<^in<or^<^9^o^%r^o«'r<^«cM— CMr^NOl^<^r^i■n•o9'K)rO(neo^^o^«o(^***<ol^tcM■^tn«o^(^r^1r 



•- in CSJ CM K* * — <«• m o^ <*■ <M «»• •- 



<«•*— m** 




395 



N 



« « M Ki rxk •- « o ^ r^ « «- « « lA (A » « rn i«) « (A « r>k r« «> *iv «> r<k lA r«> « « » lA «- » lA CM o e m r^ » ^ ^ ^ 9" >« (O <M 



rot<)r<krM«oiA«oers.iA<ro^er^«'^«'^tf>^<«o^tA«OK)K)fga««^^9^<rN'^^»«0rvin«roiAh^f4^«oCMiA 

•Ocgmar^aK1tA^r^or>9><tf^*-^^4no»K)<^«r^J*'r^•tno«l0^e^lnopor^e■o<or«»o^^Ni0^r^c^J««-iOK) 

cJH >o9«e^(^io^NiAr^^9^ioiA^iAN^o*oio«-fO««-^fO^N^(Moo»oitf'0<^in^^«3amoior^N*-r*kiA 

»^^KlO^^^^^A^r•«o^<o^^^aN^*<^»nM•-p^^J'*'•o«o•nn'^r*o>olOo^*^K^«K»M*{MoK»^^M<M♦ 
ofO ^^^i^N ^tri<«r^»OfO«0N^N^ •«<ctAK)rMiA^NiA tf>p^xoiA NiA^m^io (ooo^ r^ra^<or4 



aJ ^ecvieoiA^«o^ee**o— aooooe^ooec 



*^r<>«rs.OCMlA9«i0«M^«0K1«>*-^«|<«-(MOK^(Ma0»«-«-9>fM(Ar>ke^0»»9>lA9«lA<«»OP0t0f»^^tfCNjr4OOr<.^ 
^«MC^~"'- ---- --__ -__- ^____ __ 



aa<tf'r>j(MP^r^«rs,pn<^o»9»<^fOtf>r>«i»K»'^io»iA^«9>Np^e'«^ioiocvJOCM'«iA«<^p**orN><Mao^o9>ri>N»o^ 
«oMNrr^rg^^^^CM^o^rsi^»or<h^iAfOr^(MC>jNr^^iotfiAK)^r«oioeNJi<i9^o^<iOK)>*in^o^e^iAcgN 

CM*- CM '-KICM ^CM^ r^CM^^'-CM <OtA<OfO CM ^ O lOCM lO^^r* ^ ••rtfti r*^ N 



te^NiO•OiOKt^AK)^^o«o^^^^iO^<i'«•^^<^CM^•oo^^e^lOom^o^u^cM^«^«9>^NOln>oiO^^<^CM(A« 

^CMO*«ina»rs — fM*'CM^»O^^^CMr*^iOiON*-^<^CSJiOr*lAr>.*-^*-CJ»-0— ^^^iOCM^*-»OlO<M^lA^«f 



oo^moooooec 



>a^eo<aeooe^o*-eeMe*~'OaiA«-^oe^oev-oe«-oe^ 



<««r^e^ClllrtCM"•<^«0"«K>fM•-(^JO•-fUlAtA«09>a^*-at«^maor^>OlO^^<^CM^^CMlA(MfO<a»iQr»>«o^ 

iooi/^<^iAo^«O(^^or*o^cM^9'tf«0Oio<«Ph9*^»o*~r^o^r^<oomaooiAtncM^too^rhio^etdsooiA«Ovo 
(Oioi^Kir^r^o«}i>oeoo^iA^^oto(?>^Okioo^0^r^9t^r^^fOforMr«9to»9>ir)CMix^<Mr<>iAo«o^e^«Oior*>CM 

sf*^%r'*feM^^>r^^ifiofsjocrff'«or^*CM^ff'<7"*''*'0^r^^<^*"0'«»omwtn^o^rt^iAf^oo^ortCMo«o 

Kl «m**K1CM CM^ ^^ ^ ^ *K» ^rt ^ »-^>*^ — ^r^K>"- «* (ACM — * M 



oe 

>• 
n 

z 

CM O 

4> I- 

rH < 

^ -J 

H a. 



»9>CsJf^^«CMroM^«O^roo«ir)mmiAttiAaaiA*ooo<?»<^(MtAor^iAr«>oK>seror><o(M^o^r^«*^«O^Ki%0^<^»< 
H* Nom^r^ro^m9'^car^«oocM^(n^K>ootf-'or^«OO^CM^<^No<ir<rinr^r<k^foa»r^^«oaOK>ror<^o«^CMflOiAiOiOr 

o 

04 ^(ArocM^r%to%o«iooKi^*-a^rvinf^(^ecMiocM*-ino^>o^ocM^fONoo«<«^MNoaiA<-oe«OCM>OfO< 

'-CM «- ^r**'^ ^ CMN^W^ •-^CM KJ^** •- 

0^K>ia(^tnoln«-r^e^lAr^^^<^«0«0Q^CM'M9«r^lA•-lnM^K>tr(^^i0'^^CMotfM<-»rM•0^*-«)ln^«-c^< 
•*-«o^iA(M*off»^iArs,iOK>«*"»o«(M»n«or.«o«-fMr^<M^tOrt>rocjor**-inr>*'T^iA^^'omo'*'0^«0^o*^p 
rt(Mr>*^o<o^^^K»^^»i>r*^*rCMiocM*r'0^o^K>^ior*o*(MK»o*oeoo*r*«o^iOioff»^iOP^>«^^^o^r 

CM»«iO'*Ki«»->o^«io«Tr^CM*'r<M^«»«ior*.— rN.cMrwto^fs.^*^rtif»o*f>j^^fs.if»rsir*.CM^^a''-«iAiAKiii 

lACM^in««- «-IA 
O^CM*- «IA 



CM 

^9^K)iA^»^^^r*>«Oo^r^^io^^<^^^eQiACM^o«raO«f0^rN>^K)^^r>«>^^CMoeKie<^PO<^K)^iA«0» 
K)ior<>*ocM^9'ecMrorocM^iA^*-r<^^Ne^«oirifO(n^^iO<^iArsiO'O<«9>iAiA«o^^a^iri«ocMKi<r^<^^9^<0^tA 



Of 



fM9>«iOtAr<^afMCJlnlA*^lAlA^•.^•<0^«er^c^JKlMln(MCM•-e*o^forsi<TOiO*OK)««««lAo«^r^CMmo««» 
^ ao>oiOi«^irtr>>c^oeirtao(M^e^(Mtnn^pv<^^r»(\ji'^*-ocM(ri(n»'^m^r<h(n'o^9*^ e(M^t0trto>iA«ro«<-iOiO 
acM^CMOiOCMOor^aOiA^OtdiO'OeCMf^o^^^(Mo<«<^9>CMOtAiO<0^tAr0^oiOsr«riOlOO^o^eKio«i«e<^ 

K)r^^^«(Si^^«or^flO«fOo<o^^r*»«o*'^^r^^<r^>r«ro^a^tn^fO<0CMea^<«o^K)K)^iOOKio^r^<ir^ 
eiOlOCMiOCMU^•o>r•-*-^ro9CMa«0'-lO^^*»lO«o<^^ror>»^r^•l^^e«o<^^^^^*■oa^^<a•0'^fO^CMr^«0*T« 



o e>CM CM*- aocgcM 



■» CM M lO CM ^ lO lA r». »0 •- i*- ^ >0 «*" "T * CM CM O CM « -■ »- ^rt* 



in«o*-iAiAN^^fO>o^iA^m«o<^«o^r^oeinr^«00«]<oeiAK)ef«)^CM<«r^oainm^o«oe«-r<»^«0^<^r^p 
O oio^A^lOo^or^lOKlfMo<^K>— (Mo^^O'«^^Kl^^fs.looOff'CMO»^^sJ^rN.^^K)c^o<7*tf^(M^fiCMa'K)ln•-^n^1«M 
^1-4 a«iflorM^v>9'(AiOiOio*-^(MA(MiO^r<»^^^eo9«^«io«^«0iOaOor»r<k«CM^«O**iOr^^*-o<^«O^«r<hU 

H*^ lOrO"-iO<OP^9»r«>ira^fO^Kt«0OK>K)OlAotf'<«r^C4IAO4<O(^OOtfCM<O^CMrNinK>lOr<»*O*-^^^<«C\J<h(AO 

o^ ^9>o^iO<OiOoo^K)^^iO%r(MO««-^^^CM^'0^r^(M^iO'Oocsj^etA«oiAe^iMio<«srrM(^<^CMNO^>rio<reN 
>-3 ln«}<r^kr^J<A«o^lA«^o^o^o^«9•lO<0CM^c^lr^CMelAo^p<•lA■30•»o*OlAflO•«^^O'«•o<^^<0lACM^lAto•-e>rk^ 



Z«00 Zmi 

<-lC^^UJ OC9Z 

< >«/X<Ui «^QCM 

_ __ ___ oz> tn zzz>zz 

j<^ozu.Q£UJ3 'MOMaz^ Otrtui^<i>-<uji-«o<<<QZ'm;>-zx zotnujxxujtn oi-ti-i om 
^aa)N<MOZ<^Qeac<<XHti-K(nh-i-iz>-«nzz(n(nH-Qe< H-^e<ozOH-»-z<xcozt-oc 

■-K<i-t^-j^z-i«noa3<<^a3ZZ3i-tQCfrtuZ(n(nzA>3333oco£M^iiizo33ZX<Koe(n(n<no 
z-j^ococ<oaujM^uj<o^zo<<uJO<<<i-4)-iMMOujujiiJUJUJUJoox^oeujzooujuji-uii-i<ujM>- 

a<-<<<UUaOOlLOX>HHiiHl-i^^_lS:CSE££CCZZZZZZZZOOO^0e<AtflH>H-3>>3377 



■< < 3 O 






lu a. x>-o oo 


»- IH O 




< 


in <a. inuiu ae^ 


in tfizoMUJu. 


in 


vz 


03ZI-1-IM < a.ini-i^-«< 


« <-<oeoi-aea'<< 


►^< 


^< 


zx<o<na;-<:^ Cqcxd^uo 


oc<zino<u^ aMM 


oz 


muii 


<<jo<n<n3Zin<«uJUJO 



396 



Table 3 A 
Black Population 



1 •53% of 


Blacks live 


Ln 


the South 








1 'States with the 


largest Black 


population as percentag 


e of total 1 


1 1. 


[District 














of Columbia] 


70.2% 


7. 


Maryland 


22.7% 1 


1 2. 


Mississippi 




35.2% 


8. 


North Carolina 


22.4% 1 


1 3. 


South Carolina 


30.4% 


9. 


Virginia 


18.9% 1 


1 4. 


Louisiana 




29.5% 


10. 


Arkansas 


16.3% 1 


1 5. 


Georgia 




26.8% 


11. 


Delaware 


16.2% 1 


1 6. 


Alabama 




25.6% 


12. 


Tennessee 


15.8% 1 


1 'States with the 


largest number 


of Blacks 




1 1. 


New York 




2.41 mil 


7. 


North Carolina 


1.32 mil 1 


1 2. 


California 




1.82 mil 


8. 


Louisiana 


1.24 mil 1 


1 3. 


Texas 




1.70 mil 


9. 


Michigan 


1.20 mil 1 


1 4. 


Illinois 




1.67 mil 


10. 


Ohio 


1.08 mil 1 


1 5. 


Georgia 




1.46 mil 


11. 


Pennsylvania 


1.05 mil 1 


1 6. 


Florida 




1.34 mil 


12. 


Virginia 


1.01 mil 1 



Table 3B 
Hispanic Population 



1 'More than 60% of the Hispanic population 


lived in three 


1 
States, 1 


1 California, Texas and New York. 






1 

1 


1 "States with the 


largest Hispanic population as percentage | 


1 of 


total 










1 1. 


New Mexico 


36.6% 


7. 


Florida 


8.8% 1 


1 2. 


Texas 


21.0% 


8. 


Hawaii 


7.4% 1 


1 3. 


California 


19.2% 


9. 


Nevada 


6.8% 1 


1 4. 


Arizona 


16.3% 


10. 


New Jersey 


6.7% 1 


1 5. 


Colorado 


11.8% 


11. 


Illinois 


5.6% 1 


1 6. 


New York 


9.5% 


12. 


Wyoming 


5.1% 1 


1 'States with the 


largest number 


of Hi 


s panics 




1 1. 


California 


4.54 mil 


7. 


New Mexico 


0.48 mil 1 


1 2. 


Texas 


2.98 mil 


8. 


Arizona 


0.44 mil 1 


1 3. 


New York 


1.66 mil 


9. 


Colorado 


0.34 mil 1 


1 4. 


Florida 


0.86 mil 


10. 


Michigan 


0.16 mil 1 


1 5. 


Illinois 


0.63 mil 


11. 


Pennsylvania 


0.15 mil 1 


1 6. 


New Jersey 


0.49 mil 


12. 


Massachusetts 


0.14 mil 1 



397 



Table 3C 
Asian/Pacific Islander Population 



1 .60.0% 


of Asian/Pacific 


Islander populat 


.ion lives 


1 
1 


1 in the 


Pacific Division 


of the West 






1 
1 


1 .States with the 


largest Asian/P 


acif 


1 
'ic Islander population | 


1 as a percentage of 


total 








1 1. 


Hawaii 




61.2% 


7. 


Maryland 


1.60% 1 


1 2. 


California 




5.6% 


8. 


Oregon 


1.60% 1 


1 3. 


Washington 




2.7% 


9. 


Illinois 


1.50% 1 


1 4. 


Alaska 




2.1% 


10. 


New Jersey 


1.50% 1 


1 5. 


Nevada 




2.0% 


11. 


Utah 


1.40% 1 


1 6. 


New York 




1.9% 


12. 


Virginia 


1.30% 1 
1 


1 .States with the 


largest number 


1 
of Asian/Pacific Islanders | 

1 


1 1. 


California 




1,31 mil 


7. 


New Jersey 


0.11 mil 1 


1 2. 


Hawaii 




0.60 mil 


8. 


Virginia 


0.07 mil 1 


1 3. 


New York 




0.33 mil 


9. 


Pennsylvania 


0.07 mil 1 


1 4. 


Illinois 




0.17 mil 


10. 


Maryland 


0.07 mil 1 


1 5. 


Texas 




0.13 mil 


11. 


Michigan 


0.06 mil 1 


1 6. 


Washington 




0.11 mil 


12. 


Florida 


0.06 mil 1 



Table 3D 
American Indian Population 



1 .50.7% 


of American Indians live in 


the West 




1 .States with the 


largest Americ 


an Indian population 


as 1 


1 percentage of total 










1 1. 


Alaska 




16.0% 


7. 


North Dakota 


3.1% 1 


1 2. 


New Mexico 




8.2% 


8. 


Nevada 


1.8% 1 


1 3. 


South Dakota 




7.0% 


9. 


Wyoming 


1.8% 1 


1 4. 


Arizona 




5.7% 


10. 


Washington 


1.5% 1 


1 5. 


Oklahoma 




5.7% 


11. 


Utah 


1.4% 1 


1 6. 


Montana 




4.8% 


12. 


Oregon 


1.2% 1 


1 "States with the 


lar 


gest number 


of American Indians 




1 1. 


California 




0.23 mil 


7. 


Washington 


0.06 mil 1 


1 2. 


Oklahoma 




0.17 mil 


8. 


Texas 


0.05 mil 1 


1 3. 


Arizona 




0.15 mil 


9. 


Michigan 


0.04 mil 1 


1 4. 


New Mexico 




0.11 mil 


10. 


New York 


0.04 mil 1 


1 5. 


North Carolina 


0.07 mil 


11. 


Montana 


0.04 mil 1 


1 6. 


Alaska 




0.06 mil 


12. 


Minnesota 


0.04 mil 1 



398 



f 



Nmiaer of U.S. 


Counties With Over 20 Percent ^ti^ority Population, By State 




lUIAL 
ND. OP 




COUNTIES WITH >20 PESCENT 






AMERICAN ASIAN/PACIFIC 




OOUNIIES 


BLACK 


HISPANIC INDIAN ISLANDER 


ALABAMA 


67 


37 




ALASKA. 


23 






AFT7CNA 


14 




8 3 


ARKANSAS 


75 


27 




CAIIECRNIA 


58 




U 1 


CnrfEAEO 


63 




12 


OCNNECTICUr 


8 






lEUmSE 


3 






DISISICT OF CCEIWBIA 1 


1 




FLORIDA 


67 


13 


1 


GEORGIA 


159 


108 




HAWMI 


4 




4 


mwK) 


44 






TTT.TNDIS 


102 


4 




INDIANA 


92 


2 




KM. 


100 






KANSAS 


105 


1 




KENIIXXY 


120 


1 




LOUISIANA 


64 


46 




MMNE 


16 






MARYLAND 


24 
14 


10 




M^SSAOnjShTib 




MTCHIGAN 


83 


1 




MMESOIA 


87 
82 


65 




MLSSISSLPPI 




MISSOURI 


115 


2 




>finANA 


57 




5 


NKKKASKA 


93 




1 1 


NEVADA 


16 




1 


NEWHAMPaOEE 


10 






NEW JIESEY 


21 


1 


1 


NEW^fXIOO 


32 




28 3 


NEW Yaac 


62 


3 


2 


NCKIH CARfil.Tm 


100 


56 


2 


NdOHIMaCaA 


53 




3 


one 


88 


1 




CKLAHIA 


77 






CREGCN 


36 






PENNSYLVANIA 


67 


1 




RHMH ISLAM) 


5 






SOUTH CAROLINA 


46 


40 




SOUffl EAKOIA 


67 






TENNESSEE 


95 


8 




TEXAS 


254 


27 


70 


umH 


29 




1 


VHMGNI 


14 






VIRGINIA 


98 


47 




WASHDCTCN 


39 




1 


WEST VIRGINIA 


55 






WISCOtiiN 


72 




1 


WHMNG 


23 







399 



Thirty-seven states have at least one county with a single minority's 
population of 20 percent or greater. In eight states, over half the counties 
have a single minority's population of 20 percent. 

The number of states* having at least one county containing a "20 percent or 
more" minority group are: Black (23), Hispanic (10), American Indian (9) and 
Asian/Pacific Islander (2). Note: Some states have more than one type of 
minority county with greater than 20 percent of that minority population. 
Further, because of the relatively few counties with a 20 percent or greater 
Asian/Pacific Islander population, a lower density of 10 percent or even 5 
percent was used in some analyses. 

Minority County Population Distribution 

Of the 3,077 total U.S. counties, 1,265 (41 percent) are urban, 1,157 (37.6 
percent) are rural, and 655 (21.3 percent) are metropolitan, where "urban" 
denotes that not more than 25% of the population lives outside a city, and 
"rural" denotes that not more than 25% of the population lives within a city, 
and "metropolitan" encompasses all others. 

The counties with large minority populations are distributed among 
metropolitan, urban and rural counties somewhat differently than are the 
counties without large minority concentrations. The smallest differences 
occur in heavily Black counties. Counties with heavy concentrations of Blacks 
(20 percent or more) are predominantly urban in nature, more so than are the 
counties with smaller concentrations. Over 45 percent of the 503 counties 
with heavy Black concentrations were urban as compared with 40.2 percent of 
the counties without such concentrations. Heavily Black counties were 
somewhat less likely to be in metropolitan or rural areas than their 
counterparts. 

Among Hispanics, differences were more widespread. Counties with a heavy 
population of Hispanics were less than half as likely to be in metropolitan 
areas as low-Hispanic counties and significantly more often were found in 
rural areas. Almost half (49.9 percent) of the Hispanic counties were rural 
counties, while only 38.4 percent of the non-Hispanic counties were rural. 

American Indians were distributed in much the same way (by county) as were 
Hispanics. Nearly 47 percent of the counties with high density of Indians 
were rural, as compared with 37.5 percent of low density Indian population 
counties that were rural. American Indian counties were far less prevalent as 
either metropolitan or urban areas. 

By way of contrast, Asian/Pacific Islander population centers were heavily 
concentrated in urban areas. 

The distribution of the "high minority population counties" within these 
sub-groups is provided in Table 5. 



Includes the District of Columbia. 



400 



Table 5 

Distribution of High Minority ( 20.0%) Counties by 

Minority Group and Location 



1 
1 


Total 


Metro 

# % 


Urban 

# % 


Rural 1 

# % 1 


lU.S. Counties 
1 


3077 


655 


21.1 


1265 


41.1 


1157 


37.61 
1 


1 

1 Black 

lAll Other 


503 
2574 


99 
556 


19.7 
21.6 


229 
1036 


45.5 
40.2 


175 
982 


1 
34.71 

38.21 

1 


1 

1 Hispanic 

lAll Other 
1 


154 
2817 


15 
640 


9.7 
22.7 


63 
1202 


40.9 
42.7 


76 
1081 


1 
49.41 

38.41 

1 


1 

[American Indian 

lAll Other 
1 


32 
3045 


5 
650 


15.6 
21.3 


12 
1253 


37.5 
41.1 


15 
1142 


1 
46.91 

37.51 
1 


1 

lAsian/Pacific 
1 Islander 
1 (Over 10%) 
lAll Other 
1 


6 
3071 



655 



21.2 


4 
1261 


66.7 
41.1 


2 
1155 


1 
1 
1 
33.31 
37.61 
1 



The variations in these statistics between the four minority groups reinforces 
their individuality and the need to review separately their communities and 
their health professional resources. Consequently the following discussion 
reviews in turn, information derived from the Area Resource File pertinent to 
the "high minority population" counties for Blacks, Hispanics, Asian/Pacific 
Islanders and American Indians. For each minority group, the discussion 
presents demographic characteristics followed by data and discussion of the 
degree to which health professionals are available to that minority group. 

In developing a descriptive picture of the minority communities, data compiled 
from the ARF which may affect health outcome were analyzed. The data selected 
for review by this group were the following: 

1 . Personal income 

2. Infant Mortality Rate (IMR) 

3. Aid to Families with Dependent Children 

4. Level of poverty 

5. "Health Manpower Shortage Area" and "Medically Underserved Area" 

6. General hospital utilization 

7. Physician manpower and distribution 

8. Distribution of other health professionals 

This section of this report is divided according to the four ethnic/racial 
groups; Blacks, Hispanics, American Indians and Asian/Pacific Islanders. It 
utilized the aforementioned selection factors to evaluate the gross health and 
economic status of minority communities and the relative availability of 
health professional resources. 



401 



The second part of this section examines data on minority health professionals 
— their development, distribution and practice. 

NOTE TO THE READER 

It is critical at this point to remember that the process for fully 
delineating the various minority communities could not be completed within 
the time constraints of this study. As a proxy measure, the working group 
has chosen instead to analyze data available, primarily from the Area 
Resource File, on those counties or groups of counties within various 
states, if they had a 20 percent population of a minority group. [The 
exception regarding Asian/Pacific Islanders was noted above.] 

There would be many flaws in attempting to directly compare events or 
circumstances in these analyses to, for example, a specific Chinese-American 
or Puerto Rican community in a given city or county. Nonetheless, an effort 
was made to proceed in reviewing the county-level statistics to see if any 
consistent patterns or trends might be identified among the several study 
variables by state, regional or national perspectives. 

It was also felt that this examination of county-level data represented a 
necessary preliminary step (just as the prior review of national and state 
data had been), in presenting what can evolve into a more definitive analysis 
based on review of other inter-county and/or sub-county level data. 

PART II. A. 1. - Black Coimnunities and Health Professional Resources 

This section of this report will focus on the Black community by documenting 
various differences between densely populated counties (i.e. greater than 20% 
of the population being Black) and the rest of a particular state. For this 
presentation, 503 counties, out of a total of 3077 U.S. counties, from 23 
states and the District of Columbia were included as the target or "key 
counties." The population of these counties and the District of Columbia 
consisted of 15,797,000 Black Americans or 60 percent of the U.S. Black 
population. 

Among States containing counties with a Black population greater than 20 
percent, there was a broad range of variance from one county in one State to 
108 counties in another State. The following States illustrate the broad 
range of Black residents: one county each in Kansas (41,280); Michigan 
(829,990); New Jersey (320,827); Ohio (340,046); and Pennsylvania (638,064), 
to such large representation as Arkansas with 28 counties (326,419); Alabama, 
37 counties (820,050); South Carolina, 46 counties (841,632); North Carolina, 
40 counties (1,104,506); Louisiana, 46 counties (1,064,228); Mississippi, 65 
counties (769,860); and Georgia, 108 counties (1,335,170). 

The states having the greater numbers of key counties are primarily "southern" 
states accounting for 487 of the 503 key counties. However the key counties 
in the remaining states still account for 6.5 million of the 15.8 million 
Blacks in all these counties. See Table 6 for the full listing of these 23 
states. 



402 



Table 6 

Black Population in Selected Counties in 23 

States and District of Columbia 





Key 


Total 


Population 




Black 




Counties 


Counties 


Key Counties 


% Black 


Population 


Alabama 


37 


67 


2,310,000 


35.5 


820,050 


Arkansas 


28 


75 


1,043,000 


31.3 


326,459 


California 


2 


56 


8,156,000 


12.6 


1,027,656 


D.C. 


1 


1 


638,000 


70.2 


447,876 


Florida 


13 


67 


1,030,000 


26.1 


268,830 


Georgia 


108 


159 


3,658,000 


36.5 


1,335,170 


Illinois 


4 


102 


5,542,000 


25.7 


1,424,294 


Indiana 


2 


92 


1,288,000 


21.9 


282,072 


Kansas 


1 


105 


172,000 


24.0 


41,280 


Kentucky 


1 


120 


67,000 


25.5 


17,085 


Louisiana 


46 


64 


2,948,000 


36.1 


1,064,228 


Maryland 


10 


24 


1,747,000 


42.8 


747,716 


Michigan 


1 


83 


2,338,000 


35.5 


829,990 


Mississippi 


65 


82 


1,833,000 


42.0 


769,860 


Missouri 


2 


115 


478,000 


44.5 


212,710 


New Jersey 


1 


20 


851,000 


37.7 


320,827 


New York 


3 


62 


4,828,000 


29.2 


1,409,776 


North Carolina 


56 


100 


3,694,000 


29.9 


1,104,506 


Ohio 


1 


88 


1,498,000 


22.7 


340,046 


Pennsylvania 


1 


67 


1,688,000 


37.8 


638,064 


South Carolina 


40 


46 


2,391,000 


35.2 


841,632 


Tennessee 


8 


95 


1,450,000 


34.4 


498,800 


Texas 


28 


254 


978,000 


25.3 


247,434 


Virginia 


47 


98 


2,447,000 


32.0 


783,040 








59.6 = 60 


Percent 


15,797,401 



403 



Demographics of the Black Counties/Communities 

In each state, a review was made of several ARF indices, to help define 
differences between the target counties and the counties which comprise the 
rest of the state. If a differential need for health resources were to be 
established between the high and low density Black communities, there should 
be some related factors which could be examined to lend supportive evidence. 

One available measure for comparison of general hospital utilization in 
"Inpatient Days." In analyzing the inpatient days per 1,000 population among 
the key counties, there was a significant difference between the densely Black 
communities and the rest of the state. 

Table 7 clearly shows this gap between the key counties and the rest of the 
states. It suggests that Blacks were more likely to spend more time in the 
inpatient hospital setting at a rate of up to two to five times that of their 
counterparts in the rest of the state. Only Florida, and questionably Texas, 
do not support this trend. This may mean that the Blacks have more serious 
illnesses or more complications related to their illnesses. 



Table 7 
Inpatient Days - Per 1,000 Population (1982) 



State 

Alabama 

Arkansas 

California 

District of Columbia 

Florida 

Georgia 

Illinois 

Indiana 

Louisiana 

Maryland 

Michigan 

Missouri 

Mississippi 

New Jersey 

New York 

North Carolina 

Pennsylvania 

Ohio 

South Carolina 

Tennessee 

Texas 

Virginia 



Key Counties 


Rest of State 


1,691 


1,248 


1,836 


1,171 


1,181 


848 


2,828 


— 


1,220 


1,486 


1,542 


864 


1,686 


1,161 


1,908 


1,106 


1,447 


807 


1,889 


635 


1,659 


1,082 


5,578 


1,162 


1,608 


1,123 


2,141 


1,154 


2,045 


1,305 


1,325 


1,097 


2,002 


1,317 


2,150 


1,313 


1,213 


1,047 


2,181 


1,320 


1,364 


1,320 


1,572 


977 



404 



From available national data, individuals of all racial and ethnic groups from 
the lowest socioeconomic levels have the highest mortality rates. However, 
the level of poverty among Blacks is most striking. For example, in New York 
State, the rate in the key counties was 23.7 percent and in the rest of the 
State 9.0; in Missouri 21.8 percent and in the rest of the State 10.8; in 
Pennsylvania 19.9 in the key counties and 9.0 for the rest of the State; in 
Illinois 13.6 in the key counties and 8.1 for the rest of the State; in 
Indiana 10.8 in the key counties and 9.0 for the rest of the State; and in 
Ohio 11.2 in the key counties and 9.9 for the rest of the State. It appears 
that the differentials in poverty rates were much closer in the Midwestern and 
upper Central States than in the Eastern and Southern States. See Table 8 for 
these data. 



Table 8 
Percent Below Poverty Level (1979) 









Different 


ial 




Percentage 






Key 


Rest of 


Point 


Percent 


State 


Counties 


State 


Differential 


Difference 


Alabama 


20.3 


16.1 


4.2 


26.1 


Arkansas 


21.0 


16.7 


4.3 


25.7 


California 


13.3 


10.2 


3.1 


30.4 


D.C. 


17.3 


— 


— 


— 


Florida 


17.9 


13.1 


4.8 


36.6 


Georgia 


18.8 


11.2 


7.6 


67.9 


Illinois 


13.6 


8.1 


5.5 


67.9 


Indiana 


10.8 


9.0 


1.8 


20.0 


Kansas 


13.7 


9.6 


4.1 


42.7 


Kentucky 


19.3 


17.2 


2.1 


12.2 


Louisiana 


20.6 


13.4 


7.2 


53.7 


Maryland 


14.4 


6.1 


8.3 


136.1 


Michigan 


14.0 


8.9 


5.1 


57.3 


Missouri 


21.8 


10.8 


11.0 


101.9 


Mississippi 


26.1 


16.3 


9.8 


60.1 


New Jersey 


17.5 


8.3 


9.2 


110.8 


New York 


23.7 


9.0 


14.7 


163.3 


North Carolina 


15.7 


12.3 


3.4 


27.6 


Ohio 


11.2 


9.9 


1.3 


13.1 


Pennsylvania 


19.9 


8.6 


11.3 


131.4 


South Carolina 


17.6 


11.5 


6.1 


53.0 


Tennessee 


17.3 


15.7 


1.6 


10.2 


Texas 


16.5 


14.5 


2.0 


13.8 


Virginia 


14.0 


9.3 


4.7 


50.5 



405 



Another reflection of the level of poverty in a community is the rate of Aid 
to Families with Dependent Children (AFDC) per 100,000 population. Table 9 
shows that the rates of AFDC were two to three times higher among the key 
counties as compared to the rest of the state. Differences in the rates of 
AFDC between the key counties varied according to their geographic designation 
of rural versus urban. 

In reviewing the 23 states, and the District of Columbia, which contained the 
key counties, the distribution of families receiving AFDC is concentrated in 
non-rural areas. There were only four states (Florida, Mississippi, Texas, 
and South Carolina) which had a greater proportion of rural population in the 
key counties for Blacks than in the rest of the state. 



Table 9 

Aid to Families with Dependent Children 

Rate Per 100,000 Population and Percentage Rural (1980) 





Key Count 


ies 


Rest of 


State 




States 


AFDC/100,000 


% Rural 


AFDC/100,000 


% Rural 


Ratio 


Alabama 


6030.9 


31.1 


2646.6 


52.9 


2.3:1 


Arkansas 


5734.0 


35.5 


2036.0 


59.3 


2.8:1 


California 


6835.2 


1.0 


5355.4 


12.7 


1.3:1 


Dist. of Columbia 


15,530 


0.0 


— 


— 


— 


Florida 


4785.0 


20.8 


2317.0 


15.1 


2.1:1 


Georgia 


5096.8 


31.0 


1786.6 


51.1 


2.9:1 


Illinois 


9234.0 


1.1 


2564.4 


31.3 


3.6:1 


Indiana 


5302.5 


2.0 


2030.4 


46.1 


2.6:1 


Kansas 


8315.2 


0.9 


2371.1 


35.9 


3.5:1 


Kentucky 


5788.2 


30.5 


4528.7 


49.5 


1.3:1 


Louisiana 


5994.8 


30.5 


2622.3 


33.3 


2.3:1 


Maryland 


9,385.9 


13.6 


1922.4 


24.0 


4.9:1 


Michigan 


12,379.5 


1.6 


5471.0 


38.6 


2.3:1 


Missouri 


14,343.0 


2.7 


2880.6 


35.0 


5.0:1 


Mississippi 


8,231.2 


55.5 


3119.1 


45.2 


2.6:1 


New Jersey 


13,631.2 


0.0 


5309.1 


12.4 


2.6:1 


New York 


10,778.2 


0.0 


4561.6 


21.2 


2.4:1 


North Carolina 


4294.9 


44.3 


1761.8 


64.9 


2.4:1 


Ohio 


6973.2 


0.3 


4317.9 


30.9 


1.6:1 


Pennsylvania 


15,870.6 


0.0 


3526.6 


35.8 


4.5:1 


South Carolina 


5679.0 


47.7 


2198.4 


40.0 


2.6:1 


Tennessee 


5666.2 


11.1 


2454.8 


52.8 


2.3:1 


Texas 


2869.9 


42.6 


2100.0 


18.7 


1.4:1 


Virginia 


4561.8 


28.3 


1779.6 


38.8 


2.6:1 



406 



In all target 23 States and the District of Columbia, the infant mortality 
rate was higher among Blacks than nonminorities. The range was 23.7 per 1,000 
for Blacks in the key county versus 9.7 in the rest of the State in New Jersey 
for a differential deficit* ratio of 1.44 or 144 percent or in Ohio 28.5 for 
Blacks in the key county and 12.5 for nonminorities for a differential deficit 
ratio of 1.28 or 128 percent to 19.8 for Blacks in the key county in Kansas 
versus 16.3 for the rest of the State for differential deficit ratio of .176 
or 17.6 percent. 

The range of infant mortality rates in Kentucky was 28.6 for Blacks in the key 
county to 18.9 for nonminorities. In the rest of the State the rate for 
Blacks was 21.3 and for nonminorities 11.9. The lowest rates for Blacks were 
in New York (19.6), North Carolina (19.7), and California (19.7) for the key 
counties, while for nonminorities it was (14.5), (12.0); and (16.7) 
respectively. 

Table 10 shows that infant mortality rates were demonstrably higher for Blacks 
in comparison to the nonminority population whether in urban or rural high 
density or lower density settings and in no case in any of the areas was it 
lower than the nonminority population. 

To this point, each of the demographic features examined was consistent with a 
picture of lower health status. The working group felt that at least so far, 
the groupings of counties as a proxy for "communities," had not obscured some 
major factors known to be relevant to health status. The effort moved on to 
the examination of health professions resources in these same counties. 



The percent that is necessary for the Black population to gain parity if 
the nonminority rate remained stable. 



407 











Ikble 10 


















Infant Mortality Rate (1980)* 












State 


Key 


Counties 


Rest of State 






Non- 






Ncn- 






Ncn- 


State 


All 


ELack 


Mmority 


Total 


Black 


Minority 


Total 


Black 


Mmorlty 


AlabauH 


15.1 


21.6 


U.6 


15.9 


21.6 


10.7 


14.0 


21.3 


12.6 


Arkansas 


12.7 


20.0 


10.3 


13.9 


20.0 


9.5 


11.3 


19.9 


16.9 


CRUfomia 


11.1 


18.0 


10.6 


11.8 


19.7 


16.7 


10.6 


16.1 


16.6 


D.C. 


25.0 


26.7 


17.8 


25.0 


26.7 


17.8 


— 


— 


— 


Florida 


14.6 


22.8 


11.8 


14.8 


21.7 


11.2 


14.5 


23.0 


11.9 


Georgia 


14.5 


21.0 


10.8 


16.2 


20.8 


11.8 


10.8 


22.6 


9.5 


minois 


14.8 


26.3 


11.7 


17.3 


26.2 


12.4 


12.3 


26.9 


11.2 


Indiana 


11.9 


23.4 


10.5 


14.8 


23.5 


11.5 


10.8 


23.1 


10.3 


Kansas 


10.4 


20.6 


9.5 


17.9 


19.8 


16.3 


9.7 


21.1 


9.1 


Kentudcy 


12.9 


22.0 


12.0 


21.0 


28.6 


18.9 


12.6 


21.3 


11.9 


Louisiana 


14.3 


20.6 


10.5 


14.6 


19.8 


10.3 


13.6 


25.3 


10.9 


lyferyland 


14.0 


20.4 


11.6 


18.1 


21.5 


14.5 


10.7 


16.5 


10.2 


MLchigaii 


12.8 


24.2 


10.6 


17.2 


25.8 


10.8 


U.3 


21.1 


10.6 


MLssissippi 


17.0 


23.7 


11.1 


18.3 


23.9 


11.3 


13.2 


22.4 


10.8 


Missouri 


12.4 


20.7 


11.1 


18.1 


22.9 


12.1 


11.7 


19.2 


11.0 


New Jersey 


12.5 


21.9 


10.3 


16.4 


23.1 


9.7 


12.0 


21.3 


10.4 


New York 


12.5 


20.0 


10.8 


16.0 


19.6 


14.5 


11.0 


20.5 


9.8 


North Carolina 


14.5 


20.0 


12.1 


15.0 


19.7 


12.0 


13.5 


21.6 


12.4 


Ohio 


12.8 


23.0 


11.2 


17.7 


28.5 


12.5 


12.0 


20.6 


11.0 


Pennsylvania 


13.2 


23.1 


11.9 


19.3 


22.9 


15.8 


12.1 


23.4 


U.5 


South Carolina 


15.6 


22.9 


10.8 


16.8 


23.8 


U.l 


U.l 


15.1 


10.1 


Tennessee 


13.5 


19.3 


11.9 


15.0 


19.9 


10.9 


12.7 


17.5 


12.2 


Texas 


12.2 


18.8 


11.2 


14.3 


20.0 


11.8 


12.0 


18.6 


11.1 


Virginia 


13.6 


19.8 


U.9 


15.0 


20.0 


12.1 


U.7 


19.0 


11.7 



*(Deaths per 1,000 live births) 



408 



Availability of Health Professionals 

In examining the availability of health professional resources, it was 
apparent that more information was available regarding physicians than any 
other health providers. For the sake of brevity, much of the discussion which 
follows will focus on that data, with reference to other health disciplines as 
time and space permit. The working group did not feel this to be 
inappropriate since the methodology was still evolving. A more equitable 
presentation by discipline is anticipated to be forthcoming. 

To review the availability of physicians to minority communities, a reference 
point was needed. Since the county data was based on the 1980 Census, it was 
felt appropriate to use the national physicians per 100,000 population ratio 
for 1981 as an index. That figure was 186 . 

In 12 of the 23 states the ratio of physicians per 100,000 population in the 
key counties was lower than the national figure. Generally these were the 
same states previously described as "Southern," containing the largest number 
but also least densely populated of key counties. Where the ratio exceeded 
the national figure, a cursory review of the states suggested a correlation to 
the location of medical schools. This could significantly overstate the 
numbers of physicians assumed to be available directly for patient care. 

A review of physicians per 100,000 population indicated that in all but four 
of the 23 States (District of Columbia excluded), the number of physicians per 
100,000 population were greater in the key counties than in the rest of the 
State. These exceptional States were Florida, Kentucky, Michigan and Texas. 
In the other States the number of physicians per 100,000 population in the key 
counties ranged from 103 in Texas and 108 in Georgia to 402 in New York and 
635 in Missouri. 

When comparing the number of Black physicians per 100,000 Black population 
(excluding California, where insufficient records are available, and Kentucky 
where the rate was too low to compile), the range was 7.8 per 100,000 
population in Mississippi to 16.2 in South Carolina to 69.3 in Illinois, 70.5 
In New York and 105 in the District of Columbia. In none of the States except 
the District of Columbia did the ratio of Black physicians per 100,000 Black 
population exceed 40 percent of the national ratio for all physicians to 
population. See Table 11. 

There were no generally consistent patterns observed in the change in 
physician population comparing the key counties to the rest of the State, 
across various states, between 1970 and 1980. The increases or decreases were 
not consistent with the population growth nor its decrease. 

For example in Alabama, the physician population grew by 69 percent In the key 
counties and 66.3 percent in the rest of the State, which indicates similar 
growth. In Florida, physician population Increased by 81.5 percent in the key 
counties and by 109 percent in the rest of the State. 



409 



Table 11 

Physicians Per 100,000 Population 

in the 23 States & District of Columbia 





All 






Black 






Physicians 




Physicians 










Per 100,000 


State 


Key Counties Rest 


of State 


Number 


Black Population 


Alabama 


159 


75 


219 




27.2 


Arkansas 


153 


94 


65 




20.4 


California 


286 


198 


— 




— 


Los Angeles 


1,123 








121,3 


San Francisco 


119 








141.1 


District of Columb 


La 362 


— 


467 




105.0 


Florida 


150 


183 


61 




24.2 


Georgia 


108 


75 


442 




33.9 


Illinois 


243 


126 


980 




69.3 


Indiana 


223 


99 


80 




28.6 


Kansas 


320 


136 


14 




33.9 


Kentucky 


113 


212 


NA 




NA 


Louisiana 


177 


99 


195 




19.0 


Maryland 


333 


218 


443 




60.8 


Michigan 


148 


169 


479 




58.1 


Mississippi 


113 


73 


60 




7.8 


Missouri 


645 


103 


98 




46.4 


New Jersey 


293 


168 


210 




67.4 


New York 


402 


207 


937 




70.5 


North Carolina 


163 


130 


292 




26.7 


Ohio 


334 


155 


203 




60.1 


Pennsylvania 


285 


134 


181 




28.6 


South Carolina 


130 


107 


128 




16.2 


Tennessee 


268 


97 


261 




52.6 


Texas 


103 


148 


76 




31.4 


Virginia 


173 


157 


233 




31.1 



410 



In Georgia the physician population grew by 61.4 percent in the key counties 
and by 102.5 percent in the rest of the State. In Illinois, from the 
physician population in the key counties grew by 34.7 percent compared to a 
growth of 66.7 percent in the rest of the State. In Louisiana, the population 
m the key counties grew by 47.5 percent and by 117 percent for the rest of 
the State. In Maryland, the growth of the physician population in the key 
counties increased 31.6 percent, and increased by 136 percent in the rest of 
the State. 

Other Health Professionals per 100,000 population: Examination of the 
distribution of health professionals in key counties (greater than 20% of 
population was Black) in all the twenty-three states and the District of 
Columbia showed a higher rate of health professional per 100,000 population 
than the rest of the state. The exception to this finding was the State of 
Kentucky. See Table 12. These figures for pharmacists, registered nurses, 
dentists and optometrists follow the trends of the physician data and are also 
felt to be influenced by the locations of their respective educational 
institutions. 

The raw data necessary to investigate this and related questions has been 
compiled and will be analyzed in the near future. 



411 



a 






10 






u 



pi w 



u 



1^ 



O -u 



ti 



II 



■g a 

CO 



I" 



■8 S 



I- 



u 



1^ 



^ o 



cx) m o 



d d 



00 m m 



vO o 



3 



ON Lo 00 ^£) r^ 



vO o 



r^ 00 u~i 



d ;=l 



o 00 m vo 






oom cNvO(r)inrHr~.cNoo 



J^ 



IT) lO I m 



^ S 5»; 5 5! 



ro in 



R 



OS vO u-| 



.H\D0QO>r-!r~>-*Q 
r-.strO<rrO-^roroin 



m 

i-i 

CM 



CO CN . f^ -^ 



CN m 



<■ CN 



ro 1.D r^ m o in 

Q 00 On CO tH m 

in CO -d- en CM ~a- 



??; 



ON CO r-l CO Q ON ON 

rH -d- I o m CO m vo 
ro <• >* ~d- m m >25 



in 

CM 



en 

CTN 
CN 



OQ O 'i) vO CD 
vS iH ON 00 v5 

cN<roo-3Fcnin-^incncN-* 



t^ 00 o r^ CN m 
CN rH en cjN r^ 



s 



St ON ON v£) go 
in CM I en m en 



in in i^ ON CM 
~d- -* -d- CN m 



^ S en 



00 <r m r^ 
en -* -d- r- 



s 



f^ 'd -d 
en <f ^ 



oooQQ<i-ONCNiH<^siinr^r:i>g: 
ininenr-~m~d-vj-in<)-~a-en-* 



m >;}■ 00 en m vD ON 
r~- ON ON I 00 r^ CM ON 



~d- <N g\ 00 

-9- en On iH 

i-l r-J csl 



ON en o 

a a s ' 



s 



-a- m en ON -a- 00 en 
r^ >a- CM <3N H c^ en 
iH CN CN CM H tH en 



in i-H ON 

CN CN CN 

1— 1 r-l CN 



ON rH en 00 
r~- <r 00 -• 



__ m CN m vo 
CM in en m 'S 

H iH H .-H CN 




412 



& 



u 

4) K 



u 



i;o| 

.•^ 

H 
^ 

4-1 4-1 

(2 M 



tl 



13 



to 






^ 



CO 



•H 
P>^ 4J 



ly to 

O 4J 
CO 



u 



1^ 



t-jr^o^',Dr^0^o^OO^oo^~~vo 



00 vo 



^ a\ r^ m 



a 



vO r^ 



Ooo\^oovoooor^cj>t^r^ 



s 



CNPOmvOrO-^O-rOrOm-5 






o^ ^ o <r c<n 
eg CO r^ r^ ro 



~* ^ Q vO '•O -<t 
vo Lo c'^ in m >3- 






00>d;Ql^mmQiHCMr^CNl 
csj>3-Ci5v£>ro-<t-i7Sm~d-ro^ 






c^^Qo^cr)nooc3^Hc^•^"^ 

LOCFivOvOvOt-^i-ICN- - 

csi ro ~a- Ln 



vo t-~ iH cnI r< i-^ o> 
m ^ v£> CO CM ro ro 






Q Q ^r CO ~a- 00 <f 
00 ^ r^ t^ r~. <j\ CJ^ 
CM 00 m lO <r m un 



o^ vJS Ss 
~a- CM ~d- 






P~.->t00v5cOCy\r-|in-*rHCN 
CSJ^-*l7S-^-*vDCOCOCO-<)- 






s 



mcoo>vor-~mrHr^co-4 
u-i<rco~*co~d-^-<f<f-T 



CO 



iHCNJ<fOOOCNvD<f~g:Q 

aNm<f<tcor^tovo~T'Ji 






r^(yi-^i— ivpr^ooi^cMcor^ 
LO in ^ St •S- CO ^ 



tn in -* ~d- 



^rHC^iHrHr-li-lrH.-liH 






i^inoomcooHgocsiin^ri 



j3--lr-ICMiHi-l--(r-|r-liHr-( 



&^ 



CO 



a 3* a ;5 i:^* s 9 8 J^ ?^ ?5 ?5 



413 



PART II. A. 2. - Hispanic Communities and Health Professional Resources 

Inferences about health facilities and manpower resources available to 
Hispanic populations must be made cautiously. The basic area profiles of the 
Bureau of Health Professions' Area Resource File (ARF) provided basic data on 
health facilities, manpower resources, population, and income of Hispanic and 
non-Hispanic areas. These were supplemented by data on the size of the 
Hispanic population and the aggregate number of Hispanic physicians available 
in the target counties. However, in the absence of Hispanic-specific data on 
incomes, rates of utilization, numbers of other health professionals, and the 
like, one cannot necessarily conclude that Hispanics have the same access to 
resources or use these resources to the same degree as the general population 
of the target areas. This is particularly true since Hispanics are a 
minority, although a sizable one, in all but one of the areas examined. 
Further, there was no information in this data base on the extent to which 
these Hispanic populations and health professionals were of Mexican-American, 
Puerto Rican, Cuban, or other Hispanic origin. Therefore, no judgements have 
been made about the degree to which Hispanic communities have access to health 
professionals with similar ethnic and cultural backgrounds or whether this is 
even a significant consideration. 

Following the process outlined earlier, the effort was continued in comparing 
the key counties and the rest of the state for the high density Hispanic 
counties. Because there are only nine states in the U.S. having counties with 
a greater than 20 percent Hispanic population, some discussion of each of 
those states is provided below. 

Florida, New York and New Jersey, contain Hispanic counties which are 
overwhelmingly urban; California, has urban, rural and mixed areas; the 
southwestern and western states of Arizona, Colorado, New Mexico, Texas, and 
Washington contain primarily rural counties. 

Few common factors cut across each of these communities; perhaps the most 
striking is that except for Florida (Dade county), the Hispanic communities 
have much lower Hispanic-physician-to-Hispanic-population ratios than the 
overall physician-to-population ratio for the county or group of counties in 
which they are located. Some Hispanic communities, both urban and rural, are 
located in areas that are relatively richer (in terms of per capita income, or 
health-professional-to-population ratios, or number of health professions 
schools) than the non-Hispanic areas of their states; other Hispanic 
communities are located in areas that are relatively poorer by these 
measures. See Table 13. 



J 



414 



(-1 O • r^ 

Pj Q p ON 
O Pj iH 



:^ 



[^ L?i 



m 00 
ON 00 

00 <f 



00 a^ 



00 CNl 



CM 0-) 



?d 



1^ 

in Lo 



CO LO 



D P-i ,-1 



CM 0\ 
u-| CO 



s 



s 



ss 



~3; CM 
CO <r 



CO c-^j 



CM -5 



CO 




8 



^J o • o\ 

P-i Q O ON 

O Ph .-H 



gSR 






i-H CO 



m CO 

CO 00 



CM 00 



SR 



^S 



S 



CO VO 



to 

. . Q a! 
w • p^ 



a 0.00 
A o a\ 
tn Pj ,H 



o 



00 



o 



CTn 
CO 



CO 

m 



CO 



CM 



l-i O • a^ 
0) » o. r^ 

P-i Q P CTi 
Q Pj r-l 



^ 



^ 



u-l CM 



2RS 



r^ CO 
csj CM 



00 



s 



C^ CM 






as 



vD CM 
^ 1^ 



§^ 



Oi 



5^ H ^ 

0) > r~~ 

> Oi ON 

O M i-H 

PL, .^/ 



CM CO 



CO t-t 

c-o i-l 



o 00 



rH O 

co' O 



CO 00 



O CM 
.H CM 






o m 



CM 00 

o\ 



a 



^ 



pj 






O ON 

t-l 



<r 00 



CM 
CO 
00 



?R§8 

ON ON 



^ 00 
CNJ CO 

r^ CM 



00 ON 
00 ON 



LO ON 



\D ON 
ON CM 



ON^ 



q o 

rH rH 



00 ON 

-6«--e«- 









1^ O 



^ CO 
O cnI 



O O 

rH rH 



(2 



to S < 

0) • 1 



ON LO 
r^ rH 



CNJ CO 



00 ON 



CM 3 



?d CM 



O v£5 



CNJ CO 

CO r^ 



0) 



y 



P ON 



^O ON 



^a 



P( 



CM -^ 



LO LO 



CM LO 
CM ^ 



ON 00 
CM CM 



00 ^ 



O rH 

CM -d- 



CO O 

3?i 



1^ On 

LO CJN 



Pn 



LO CM 

CO 00 



P (i P ON 

>-l CO P^ rH 



CO 



00 
CM 



CM 



Csl 



LO 

CM 



LO 

CO 



CM 
CM 



• CO O 

e- Q °0 

P Q ON 

Pj O rH 



CM CM 
vO rH 

rH 00 



ON VO 
LO ON 



r^ 00 

LO CD 

LO 00 



ON CJN 
S 00 



Q 00 
ON CM 
00 00 



LO 00 
O C3N 



-^ LO 

00 -J- 

00 CO 

CO O 



LO LO 

vD 



?i 



ag 




a Q- 

^ -H CO 




CO 



00 

y 

N CO 



00 o 



o 
o a 



CD 




O -H CO 

O CO g 

CJ> 



00 
CO 



So.. 

•P -H CO 

jd CO 



415 



Florida 

Florida has one county with a Hispanic population greater than 20 percent — 
Dade County, which Is approximately 36 percent Hispanic. Dade County Is 
significantly different in several ways from the other overwhelmingly urban 
areas with 20 percent Hispanic concentrations (New York, 2 counties and New 
Jersey, 1 county) as well as significantly different from rural and mixed 
Hispanic communities elsewhere in the U.S. • 

In 1980, its population was 1.626 million and 1.1 percent rural, compared to 
8.121 million (18.7 percent rural) for the rest of Florida. With 796 persons 
per square mile it is much denser than the rest of Florida considered in 
aggregate (156 per square mile) but far less dense than Hispanic target areas 
in New York (40,582 per square mile) and New Jersey (11,850 per square mile). 
The minority population was 22.8 percent for Dade County and 14.7 percent for 
the rest of the state. The 1980 Income per capita was $10,582 for Dade 
County, compared to $9,135 for the rest of the state, but the percent change 
over the preceeding 10 years was comparable (296.6 and 305.1). Percentages of 
the population receiving AFDC were comparable for Dade and the rest of the 
state (2.5 for Dade, 2.6 for the rest of Florida) but considerably smaller 
than Hispanic target areas in New York (10.8) and New Jersey (10.7). The 
percentage below the poverty level was 15.2 percent for Dade County and 13.3 
percent for the rest of Florida, slightly below the Hispanic target area in 
New Jersey (16.7) and considerably below the Hispanic target area in New York 
(23.7). 

Ambulatory visits per 1,000 population were comparable for Dade County and the 
rest of the state (1,042 vs. 1,064), although the percent change from 1970-82 
was far higher for the rest of the state (96 percent for other counties, 40 
percent for Dade). Inpatient days per 1,000 population were higher in Dade 
County than the rest of Florida (1,674 to 1,415), although the percent change 
from 1970-82 was higher in the rest of the state (60 to 39). 

Dade County and the rest of the state showed striking divergences in the 
numbers of active, non-Federal physicians available, and to a lesser degree in 
other health professionals. Dade County had 301 M.D.'s per 100,000 
population, compared to 148 per 100,000 for other counties. Between 1940 and 
1980, growth rates for M.D.'s were 1,158.1 percent for Dade County and 795.2 
percent for the rest of the state, although for the last 10 years, growth was 
higher outside Dade County (118.4 to 80.4 percent). Dade also had more 
active, non-Federal dentists per 100,000 population (52 to 39). Growth in 
real numbers was slightly faster outside Dade County (11.4 percent to 9.4). 
Employed R.N.;s per 100,000 were 541 in Dade, 474 elsewhere. Dade also had a 
higher percentage of pharmacists per 100,000 in 1980 (44 to 38) and 
optometrists per 100,000 in 1981 (9 to 7). 

Health professions schools were located largely outside the target county. 
Dade had 1 medical school (175 graduates), and 5 R.N. schools (423 
graduates). The rest of Florida had 2 medical schools (212 graduates), 1 
dental school (59 graduates), 29 R.N. schools (1,964 graduates), 1 pharmacy 
school (190 graduates), and 1 veterinary school (39 graduates). 



416 



Dade County has a substantially higher Hispanic-physician-to-Hispanic- 
population ratio than the general physician-to-population ratio for the county 
as a whole. Although 36 percent Hispanic, the county had 2,375 Hispanic 
physicians, giving it 409.5 Hispanic physicians per 100,000 Hispanic 
population. The county overall had 301 physicians per 100,000 population. 
Dade has an extraordinarily higher Hispanic-physician-to-Hispanic-population 
ratio than any other Hispanic target area in the U.S., including New York 
(89.3 and New Jersey (89.0). 

New York 

New York State has two counties with a Hispanic population greater than 20 
percent — Bronx and New York, which together have a roughly 28 percent Hispanic 
population. 

In 1980 these two counties had a combined population of 2.597 million (0 
percent rural), which is about 1/6 of the rest of the state (14.961 million, 
18 percent rural). The percentage of minorities is also nearly 3 times 
greater in the Hispanic target counties (45.4 to 15.7). Income per capita was 
higher in the target counties ($11,414 vs. $10,098). Even so, the target 
counties had a significantly higher percentage of AFDC recipients (10.8 to 5.5 
percent) and of persons below the poverty level (23.7 to 11.2 percent). 

Ambulatory visits per 1,000 population were more than twice as high in the 
target counties (2,728 to 1,244) although the percent change from 1970-82 was 
higher elsewhere in the state (71 to 18 percent). Inpatient days per 1,000 
population were nearly twice as high in the target counties than elsewhere 
(2,449 to 1,345) although this figure declined 12 percent from 1970-82 in the 
target counties while rising 15 percent in the rest of the state. 

The target counties had a far higher proportion of M.D.'s, both in aggregate 
and in all specialties except general practice. These counties had 555 
physicians per 100,000 population (compared to 209 elsewhere); and in primary 
care specialties, they had 100 in internal medicine (compared to 37), 24 in 
ob/gyn (compared to 13), and 29 in pediatrics (compared to 15). Only for 
general practice were the numbers per 100,000 comparable (21 for the target 
areas, 20 elsewhere). 

Although the target counties had far higher concentrations of physicians, 
growth rates compared to the rest of the state were much lower. In real 
numbers, physicians declined 12.1 percent from 1940-80, up 2.2 percent in the 
last 10 years in the target counties. Elsewhere, physicians increased 148.2 
percent in the last 10 years. The target counties also maintained higher 
numbers of other health professionals per 100,000 for dentists (90 to 63), 
R.N.'s (711 to 540), pharmacists (53 to 39) and optometrists (11 to 7). Both 
communities showed declines in the real number of pharmacists and optometrists 
in recent years. A slight decline in the number of R.N.'s (although growth in 
numbers per 100,000) occurred in the target counties between 1972 and 1977. 

The target counties had considerable concentrations of health professions 
schools: 6 medical (926 graduates), 2 dental (233 graduates), 14 R.N. (1,346 
graduates), 1 optometry (39 graduates), and 1 podiatry (101 graduates). The 
rest of New York had 6 medical (791 graduates), 1 osteopathic (231 students, 
no graduates reported that year), 2 dental (109 graduates), 80 R.N. (5,894 
graduates), 4 pharmacy (624 graduates), and 1 veterinary (76 graduates). 

417 



The target counties had 652 Hispanic physicians and a Hispanic population of 
730,385. The Hispanic-physician-to-Hispanic-population ratio of 89.3 per 
100,000 is the highest outside Dade County, but far below the overall 
physician figure for these two counties (555 per 100,000). 

New Jersey 

New Jersey has one county with a Hispanic population greater than 20 
percent — Hudson County, which is roughly 26 percent Hispanic. 

In 1980, its population of 557,000 (0 percent rural) was roughly 1/12 the size 
of the rest of New Jersey (6.808 million, 11.9 percent rural) and far denser 
(11,850 vs. 911 persons per square mile). It had a larger percentage of 
minorities than the rest of the state (22.6 to 15.9), lower per capita income 
($9,832 to $11,088), a larger percentage of AFDC recipients (10.7 to 5.9), and 
a greater percentage of persons below the poverty level (16.7 to 8.7). The 
number of ambulatory visits per 1,000 population were similar (1,237 in Hudson 
County vs. 1,211 elsewhere), although the growth from 1970-82 was much higher 
in Hudson County (181 percent vs. 79 percent). Inpatient days per 1,000 and 
1970-82 growth rates were slightly higher in Hudson County (1,359 and 23 
percent) than in other New Jersey counties (1,260 and 18 percent). 

Except for primary care where there is approximate parity, Hudson County lags 
behind the rest of New Jersey in M.D.'s and M.D. specialists per 100,000 
population. The target county had 158 M.D.'s per 100,000 (compared to 184 
elsewhere) and in primary care specialties. 

In Hudson County between 1940 and 1980, the number of physicians grew 5.1 
percent and the number per 100,000 population grew 23.1 percent, with larger 
increases in the past 10 years (15.2 and 26.1 percent respectively). Growth 
was more significant elsewhere in the state: the number of physicians 
increased 178.6 percent from 1940-80 and 45.2 percent in the past 10 years 
while physicians per 100,000 population grew 43.6 percent overall and 39.9 
percent in the past 10 years. The rest of the state had greater numbers per 
100,000 of other health professionals, in some cases by margins of 10 to 20 
percent. Both areas have seen declines in real numbers of pharmacists and 
optometrists in recent years, comparable increases in real numbers of R.N.'s, 
and increases in numbers of dentists. 

Health professions training occurs almost exclusively outside the county. 
Hudson had only 3 R.N. schools with 63 graduates, while other New Jersey 
counties had 2 medical schools (229 graduates), 1 osteopathic (117 students), 
2 dental (156 graduates), 36 R.N. (2,065 graduates), and 1 pharmacy (104 
graduates) . 

Hudson County had 122 Hispanic physicians and a Hispanic population of 
145,249. Its Hispanic-physician-to-Hispanic-population ratio (84.3 per 
100,000 was just slightly behind that of the target counties in New York. 



418 



California 

California has 11 counties with Hispanic populations greater than 20 percent, 
and which together have roughly a 27 percent Hispanic population. 

Demographically somewhat different than the other 47 California counties, 
these 11 are very similar to the rest of the state in availability of health 
manpower resources. In 1980, the 11 target counties had a population of 9,849 
million, 6.2 percent rural, compared to a population of 13.819 million (1.4 
times larger) for the rest of the state, which is 10.5 percent rural. The 
target counties are more densely populated with 250 persons per square mile, 
compared with 118 per square mile elsewhere. The two areas were comparable in 
income per capita ($10,938 for the 11 counties vs. $10,988), in percent change 
of per capita income over the past 10 years (299.3 vs. 306.4), and in 
percentage of minority population (29.6 vs. 28.3 percent). However, the 
target counties had a larger percentage of AFDC recipients (6.8 to 5.2) and 
percent below the poverty level (13.1 to 10.0). 

The target counties were slightly lower in ambulatory visits per capita (1,173 
to 1,278) but growth in visits per capita was slightly higher from 1970-82 (26 
to 19 percent). These 11 counties had both a larger number of inpatient days 
per 1,000 (1,022 to 921) and percent growth from 1970-82 (9 to 5). 

Physician-to-population ratios were virtually identical for the two regions. 
Comparing the target counties to the rest of California, M.D.'s per 100,000 
population were 217 to 218. 

Growth in real numbers of M.D.'s has been substantially higher outside the 
target counties: 348.7 percent between 1940-80 with 208.3 percent in the last 
10 years for the target counties, 540.7 percent with 295.9 percent in the last 
10 years for the rest of the state. However, when adjusted for population, 
these growth rates are much closer, 60.6 percent for the target counties and 
56.8 percent for the others in the last 40 years, 31.4 percent and 29.5 
percent in the last 10 years. The rest of California had a higher ratio of 
dentists per 100,000 population (60 to 50) and employed R.N. 's (432 to 380 
with indistinguishable growth rates in recent years. Professional-to- 
population ratios were virtually identical for osteopathic physicians, 
pharmacists and optometrists. 

Both areas have considerable health professions training resources. The 
Hispanic target counties had 2 medical schools (314 graduates), 1 osteopathic 
(162 graduates, 2 dental (236 graduates), 35 R.N. schools (2,194 graduates), 
and 1 pharmacy school (128 graduates). Other California counties had 6 
medical schools (680 graduates), 3 dental schools (305 graduates), 51 R.N. 
(2,902 graduates), 2 optometry, 2 pharmacy, 1 podiatry, and 1 veterinary 
school. 

The Hispanic target counties had 1,055 Hispanic physicians and a Hispanic 
population of 2,697,924. Thus, while the 11 counties had one of the highest 
overall physician-to-population ratios for Hispanic areas in the U.S. (217 per 
100,000 population), they had one of the lowest ratios of Hispanic physicians 
to Hispanic population (39.1 per 100,000). 

419 



Texas 

Texas has 90 counties with Hispanic populations greater than 20 percent. In 
fact, 51 percent of the total population of these counties is Hispanic — the 
highest of any of the target communities. This is the poorest Hispanic target 
area in per capita income, and except for the small group in Washington state, 
is the poorest in terms of available health resources. 

In 1980, the target area had a population of 3.884 million compared to 10.345 
million (2.6 times greater) for Texas' other 164 counties. Classed as 19.6 
percent rural, the target area is only 1.2 times smaller in land area than the 
rest of Texas and had an overall population density of 32 persons per square 
mile, compared to 73 per square mile elsewhere. The minority populations were 
similar — 19.3 percent for the target area and 20.3 percent elsewhere. 

The similarities largely end here. Income per capita was $7,605 compared to 
$10,349 elsewhere, although growth rates were comparable from 1970-82 (336.0 
vs. 344.4 percent). The AFDC population was 3.3 percent compared to 1.7 
percent elsewhere, but the percent below the poverty level was 21.5 percent 
compared with 12.1 percent elsewhere. Only New York (23.7 percent) had a 
poverty level population within 4 percentage points of this figure. 

Much medical care in the target area appears to come on an ambulatory basis. 
Ambulatory care visits per 1,000 population were 1,492. While lower than in 
Arizona or New Mexico, this was substantially higher than the 952 reported for 
other Texas counties. The percent growth in ambulatory visits from 1970-82 
was 146 percent, nearly double the 74 percent for other counties. Inpatient 
days per 1,000 were 1,214, compared to 1,341 elsewhere, with 12-year growth 
rates in these figures comparable for the two areas (30 and 32 percent). 

In 1979, the 90 Hispanic counties had 119 M.D.'s per 100,000 population, 
compared with 155 per 100,000 for other Texas counties. This rate was below 
that of the Hispanic target area in Colorado (129) and well below the target 
areas in New Mexico (149) and Arizona (181). While M.D. general practitioners 
per 100,000 were nearly equal in the 90 counties and elsewhere in Texas (23 
vs. 24), the target area was well below the rest of Texas in all other 
specialities. 

M.D. growth rates in real terms from 1940-80 and 1970-80 were lower for the 
target area than for the rest of Texas, although M.D.'s per 100,000 grew 
slightly faster in the target area between 1970 and 1980. The Hispanic 
counties had lower health-professional-to-population figures than the rest of 
Texas for osteopathic physicians (4 per 100,000 vs. 8 per 100,000), 
pharmacists (38 vs. 45), R.N.'s (256 vs. 331), dentists (29 vs. 40) and 
veterinarians (16 vs. 23). 

Relatively little health professions training is conducted in the target area, 
which has 1 medical school (137 graduates), 1 dental school (129 graduates), 
and 14 R.N. schools (795 graduates). Other Texas counties had 6 medical 
schools (801 graduates), 1 osteopathic (328 graduates), 2 dental (252 
graduates), 42 R.N. (2,758 graduates), and 1 optometry, 3 pharmacy, and 1 
veterinary schools. 

420 



Although the Hispanic counties had only 119 physicians per 100,000 population 
overall, they did have 1,238 Hispanic physicians for a Hispanic population of 
1,963,334. The Hispanic-physician-to-Hispanic-population ratio of 63.1 per 
100,000 exceeded that of Arizona, Colorado, New Mexico, and California. 

Colorado 

Colorado has 12 counties with Hispanic populations greater than 20 percent; 
overall, these counties are about 35 percent Hispanic. In terms of income 
differences between the target area and the rest of the State and in terms of 
limited availability of health professional resources and training 
opportunities, this area is more like the Texas target area than those of 
Arizona or New Mexico. 

In 1980, the 12 target counties had a combined population of 225,000, 32 
percent rural, with a population density of 11 per square mile, compared to 
2.665 million elsewhere, (18.3 percent rural) with a population density of 32 
per square mile. The target area had a 15.7 percent minority population 
(compared to 9.9 percent elsewhere), a substantially lower income per capita 
($7,946 vs. $10,269), a larger percentage of AFDC recipients (6.1 vs. 2.3), 
and larger percentage of persons below the poverty level (16.0 vs. 9.5). 

Ambulatory visits per 1,000 population were lower in the 12 counties (1,147 
vs. 2,279) with a lower percent change from 1970-82 (132 vs. 239). Inpatient 
days per 1,000 population were higher (1,684 vs. 1,123), but the percent 
changes from 1970-82 (16 percent) were identical for the two areas. 

Except for general practitioners, the Hispanic target area had consistently 
lower physician-to-population ratios. Overall, the 12 counties had 129 M.D.'s 
per 100,000 population (compared to 196 for the rest of Colorado). 

Physician supply per 100,000 has grown somewhat more rapidly in the target 
areas than in the rest of the state. Between 1940 and 1980, the active supply 
of M.D.'s grew 48.5 percent (compared to 255.5 percent elsewhere) with a 39 
percent increase from 1970-80 (compared to 57.3 percent elsewhere). Yet the 
supply per 100,000 population in the target area was up 138 percent over 40 
years (compared to 21.9 percent) and 34.2 percent over the past 10 years 
(compared to 17.3 percent elsewhere). The target area had more pharmacists 
per 100,000 population (56 to 47), but fewer osteopathic physicians (3 to 10), 
dentists (44 to 55), R.N.'s (521 to 597), optometrists (7 to 9), and 
veterinarians (23 to 35). Both regions have had a decline in pharmacists in 
recent years. 

Very little health professions training is conducted in the 12 target 
counties: only two R.N. schools with 64 graduates. Elsewhere in Colorado, 
there are 1 medical (133 graduates), 1 dental (21 graduates), 10 R.N. (708 
graduates, 1 pharmacy (62 graduates), and 1 veterinary school (121 graduates). 

The target counties had 20 Hispanic physicians for a Hispanic population of 
79,603. Thus, in addition to having one of the lowest physician-to-population 
ratios among the targeted states, this area had the lowest Hispanic-physician- 
to-Hispanic population ratios, 25.1 per 100,000. 

421 




Arizona 

Arizona has eight counties of its total 14, with Hispanic populations greater 
than 20 percent. Together, these counties are about 25 percent Hispanic. 

Both target counties and other areas have undergone substantial population 
growth in the past 10 years. In 1980, the 8 target counties had a total 
population of 890,000, 22.4 percent rural, with a population density of 21 
persons per square mile. Minority populations were 18.4 percent for the 
target counties and slightly lower (16.1 percent) for the rest of the state. 
The target area had a lower per capita income than other Arizona counties 
($8,126 to $9,238), and a slightly higher percentage of AFDC recipients (2.0 
to 1.7) and persons below the poverty level (14.3 to 12.8). 

Ambulatory visits per 1,000 population were higher in the target counties in 
1982 (2,026 to 1,694) although the rate of growth between 1970 and 1982 was 
lower (66 vs. 166 percent). Inpatient days per 1,000 were virtually identical 
for the two areas (1,082 vs. 1,080), although the rate of growth from 1970-82 
was lower in the 8 counties (12 vs. 30 percent). 

For most specialties, the group of counties with concentrations of Hispanics 
had approximately equivalent or slightly better numbers of M.D.'s per 100,000 
than the other 6 Arizona counties. Overall, the target counties had 181 
M.D.'s per 100,000, compared to 168 elsewhere in the state. The growth in 
M.D.'s per 100,000 has been large in both areas of the state. Between 1940 
and 1980, the number of M.D.'s grew 712.6 percent in the target counties and 
1,112.4 percent outside, with both areas seeing a 92 percent growth in M.D.'s 
per 100,000. However, between 1970 and 1980, M.D.'s per 100,000 grew more 
than twice as fast in the target area (68.6 to 30.2 percent). Both areas had 
comparable numbers of other professionals per 100,000 population, including 
dentists (34 in the target counties vs. 42 elsewhere), employed R.N. 's (589 
vs. 592), pharmacists (59 vs. 54), optometrists (7 vs. 8), and osteopathic 
physicians (15 vs. 17). In recent years, both areas have shown slight 
declines in dentists, and considerable growth in employed R.N.'s. 

The state has few health professions schools, most of which are located in the 
target counties, which had 1 medical school (86 graduates), 2 R.N. schools 
(296 graduates), and 1 pharmacy school (46 graduates). Elsewhere in the state 
were 10 R.N. schools with 483 graduates. 

The target counties in Arizona had 119 Hispanic physicians and a total 
Hispanic population of 224,383. The Hispanic-physician-to-Hispanic-population 
was 53.0, somewhat higher than neighboring New Mexico, but well below the 
overall physician-to-population ratios in Arizona. 

New Mexico 

In New Mexico, 28 of 32 counties have Hispanic populations greater than 20 
percent. With a combined Hispanic population of about 41 percent, these 
counties have the second greatest concentration of Hispanics among the areas 
examined — slightly greater than Dade county but below the Texas target area. 



422 



Both regions of New Mexico had substantial population growth between 1970 and 
1980. With a population of 1.105 million in 1980 (26.3 percent rural) and a 
population density of 10 persons per square mile, the target area had about 
5-1/2 times more people than the other 4 counties, which were 36.6 percent 
rural with a population density of 16 per square mile. Both areas were 
comparable in income per capita ($7,886 in the target areas in 1980 vs, $7,996 
elsewhere) and percent change in per capita income over the previous 10 years 
(321.1 vs. 337.0). The target area and the rest of the state had low 
percentages of the population receiving AFDC (3.9 vs. 4.9), although both 
areas, especially the non-target counties, had large percentages below the 
poverty level (16.7 and 22.7). 

In 1982, ambulatory visits per 1,000 population were 1,758 for the target 
counties and 4,251 for the other 4 counties. (The latter is an anomaly; 
nothing approaching this figure was reported for any other area.) The percent 
change in ambulatory visits between 1970 and 1982 was 174 percent for both 
areas. Inpatient days per 1,000 population were higher in the target counties 
(1,016 to 832) and in the target counties, grew 20 percent between 1970 and 
1982, while declining 17 percent elsewhere. 

For most specialities, the number of M.D.'s per 100,000 population was greater 
in the target counties than in the rest of the state, including 149 M.D.'s of 
all types (vs. 92). Both areas had high growth rates of M.D.'s from 1940-80 
(453 percent in the target area and 494.3 percent outside), with higher growth 
of physicians per 100,000 population in the target area (136.6 vs. 77.3 
percent). Outside the target area, al of this growth was in the latest 10 
years (77.9 percent); during the same period, the percent change in the 28 
counties was 44.7. The target area also had higher professional-to-population 
ratios for osteopathic physicians (8 per 100,000 vs. 2 per 100,000), 
pharmacists (48 vs. 38), employed R.N.'s (383 to 322), and dentists (36 to 25) 
but a lower ratio of optometrists (8 to 12). Growth in the number of R.N.'s 
per 100,000 population was more than twice as high between 1972 and 1977 in 
the target area. 

The state has few health professions schools, most of which are located in the 
target area, including 1 medical school (75 graduates), 8 R.N. schools (415 
graduates), and 1 pharmacy school (52 graduates). Elsewhere in the state are 
3 R.N. schools with 23 graduates. 

The 28 target counties had 202 Hispanic physicians and a Hispanic population 
of 449,749, for a Hispanic-physician-to-Hispanic-population ratio of 44.9 per 
100,000. 

Washington 

One of Washington's 39 counties has a Hispanic concentration greater than 20 
percent — Adams county, which is about 22 percent Hispanic. 

Adams is a rural county in western Washington: in 1980, it had 13,000 
persons, 66.1 percent rural, with only 7 persons per square mile. The rest of 
the state had a population of 4.119 million, is 26.3 percent rural, and had 64 
persons per square mile. 

423 



In 1980, Adams's income per capita was similar to the rest of the state 
($10,030 vs. $10,237) and had grown somewhat more in the previous 10 years 
(372.7 vs. 304.8). The percentage of the population receiving AFDC was 
slightly higher than the rest of the state (4.3 to 3.7) as was the percentage 
below the poverty level (12.2 vs. 9.8). 

Ambulatory visits per 1,000 population were lower in Adams County (913 to 
1,119 but had grown more (147 vs. 60 percent) from 1970 to 1982. Inpatient 
days per 1,000 population were much lower (497 to 840) in a state that had 
among the lowest totals reported anywhere. Further, both Adams and the rest 
of the state saw declines in inpatient days per 1,000 (3 and 2 percent 
respectively) between 1970 and 1982. 

Similarly, all health-professional-to-population ratios, except for M.D.'s in 
general practice and ob/gyn, D.O.'s (the county had 2 osteopathic physicians), 
and pharmacists, were below those of the rest of the state. The county had 66 
physicians per 100,000 population (compared to 172 elsewhere). 

The county has no health professions schools. The state has 1 medical school 
(173 graduates), 1 dental school (97 graduates), 20 R.N. schools (1,141 
graduates), 2 pharmacy schools (129 graduates), and 1 veterinary school (79 
graduates) . 

The county has no Hispanic physicians and a Hispanic population of 2,950. 

PART II. A. 3. - Asian/Pacific Islanders' Communities and Health Professional 
Resources 



The caveats noted earlier regarding the difficulty in reviewing data at the 
county level for relevance to specific minority communities, need their 
greatest emphasis here. The Asian/Pacific Islanders as a minority group are 
widely spread through the U.S., with a reasonably high density in only a few 
states. Because they rarely appear even in double-digit percentages in 
county-level population figures, it is difficult if at all possible to 
attribute events seen in those counties to that minority group. 

Further, since Asian/Pacific Islanders are among the most heterogeneous of all 
U.S. minority groups, the numerous sub-groups provide even more complication 
in attempting to define a "community." For purposes of this discussion, 
Asian/Pacific Islanders are defined as being from Japanese, Chinese, Filipino, 
Korean, Asian Indian (e.g. Pakistan, India), Vietnamese (e.g. Laotian, 
Cambodian), or Hawaiian and other Pacific Islander (e.g. Samoan) ethnic 
origin. Precise figures on the numbers of each of these groups are difficult 
to obtain, thereby making it almost impossible to make meaningful conclusions 
for any of them individually. 

Nonetheless, the methodology applied elsewhere was also applied to the 
statistics for "highly dense" counties for Asian/Pacific Islanders. The 
process was modified in deference to the small number of counties meeting the 
"greater- than-0-percent" threshold, by lowering that level to 
"greater-than-f ive-percent ." [This represented a two-edged sword in that 
while providing some increase in the number of counties available for 
analysis, it reduced the already moderately low density of population even 
further. ] 

424 



According to the 1980 Census, Hawaii, California and New York contain the 
counties with the largest concentrations (five percent or more) of 
Asian/Pacific Islanders. Hawaii has, by far, the largest single Asian/Pacific 
Islander community in the U.S., however, because of its unique size and its 
political definition as a State, it was deemed to be inappropriate for 
inclusion for basis of comparison with other Asian/Pacific Islander 
"communities . " 

The problems inherent in attempting to analyze the contribution or impact of 
five percent of a county's population were acknowledged to be great. This was 
so much the case that the working group almost immediately began to look for 
alternative approaches to defining an Asian/Pacific Islander community. 

One thought was to simply recognize the reality of the dispersal of this 
minority population across the entire nation, and ignore the limits imposed by 
a five percent county population standard. For example, by arbitrarily 
lowering the data threshold to reveal counties with 0.5 percent or more 
Asian/Pacific Islanders. The population, as noted in Table 14, is dispersed 
across 51 states in 541 counties. Although the 0.5 percent in a given county 
may seem small, the actual numbers that it yields provide a larger population 
base for more specific characteristics to be drawn. 



Table 14 

U.S. Counties Sorted by Percent 

of Asian/Pacific Islander Population 



Percent 






Asian/Pacific 






Islander 


Number of 


Percent of Total 


Population 


Counties 
293 


Counties* 


0.5 to 0.99 


54.2 


1 


169 


31.2 


2 


39 


7.2 


3 


22 


4.1 


4 


3 


0.6 


5 


2 


0.4 


6 


2 


0.4 


7 


4 


0.7 


8 


1 


0.2 


10 


1 


0.2 


21 


1 


0.2 


60 


1 


0.2 


61 


1 


0.2 


62 


1 


0.2 


67 


1 


0.2 



* Fifty States and the District of Columbia contain 0.5 percent or more 
Asian/Pacific Islanders in a given county. A total of 541 counties were 
observed. 

425 



Another approach would be clustering counties that are adjacent to one another 
in a given geographic area to increase the numbers available for study. For 
example, percentages of Asian/Pacific Islanders in the Washington, D. C, 
metropolitan area may appear small when looking at individual county data. 
They only represent one percent of the population in the District of Columbia, 
two percent in Howard and Prince Georges counties (both in Maryland), three 
percent in Montgomery (Maryland), Alexandria City and Fairfax (both in 
Virginia) and five percent in Arlington County (Virginia). Each of these 
counties are geographically adjacent and together represent 86,068 
Asian/Pacific Islanders, not an insignificant figure when combined. 

Another example of the many complexities that arise when examining the data on 
Asian/Pacific Islanders, can be illustrated through the use of figures on 
health professionals. The percent distribution of Asian/Pacific Islanders 
health professionals in 1980 are provided in Table 15. 



Table 15 

Percent Distribution of Asian/Pacific Islander 

Health Professionals According to Ethnicity, 1980 



Physicians Dentists Pharmacists 



Total Asian/Pacific 

Islanders 100 100 100 



Japanese 5 

Chinese 16 

Filipino 25 

Korean 10 

Asian Indian 37 

Vietnamese 2 

Hawaiian and Other 5 
Pacific Islanders 



39 


18 


29 


41 


7 


8 


8 


10 


13 


18 


1 


3 


2 


3 



Thirty-seven percent of all Asian/Pacific Islander physicians were Asian 
Indian, while Filipino physicians make up 25 percent and Chinese 16 percent. 
The large number of Asian Indian and Filipino physicians reflects the 
influence and impact of including those who have been foreign trained. The 
ethnic composition of dentists and pharmacists are more similar where health 
professionals from Chinese and Japanese heritage better reflect the 
composition of Asian/Pacific Islanders population in this country as a whole. 

The variation among ethnic groups and where they settle, the differences 
between new immigrants and older immigrants, and the impact of being an 
American citizen for many generations, coupled with socio-economic factors, 
are just a few examples of how Asian/Pacific Islanders select their 
communities and how health service needs vary. 

426 



As a result of all these considerations, it was decided that for consistency 
with the other minority presentations, analyses should be developed for the 
States of California and New York. Those narratives follow shortly. It was 
further decided that additional analyses need be undertaken utilizing a 
variety of data manipulations including use of standard metropolitan 
statistical areas (SMSA's); sub-county and inter-county data comparisons 
across, rather than within states; comparisons of one sub-group versus 
another; etc. Several of those latter analyses are planned for future study. 

New York 

Two New York counties (Kings and New York) have populations in excess of one 
million persons each, and are urban. The Asian/Pacific Islander population in 
these two is 5.3 percent, compared to 1.1 percent for the rest of the state. 
The Black population for the two counties is 20.1% compared to 12.2 percent 
for the rest of the state (60 counties). 

In 1980, the personal income per capita for the two counties was $12,114 and 
the personal income per household was $28,396. These county figures were 
significantly higher than those for the rest of the state ($9,868 and $20,527, 
respectively). Contrary to the wealthier population base in these two 
counties, the population receiving Aid to Families with Dependent Children per 
100,000 total population was 10,777 compared to 5,220 for the rest of the 
state. The percentage of the population below the poverty level in 1979 was 
15.6 compared to only 12.4 for the rest of the state. 

These combined findings suggest significant polarization of resources. 

The Infant Mortality Rate for "other" (than nonminority or Black) is 
6.4 compared to 5.8 for the rest of the state. The mortality per 1,000 
population is 11.2 vs. 9.5 elsewhere in the state. 

Portions of both counties have been designated as Medically Underserved Areas 
and Health Manpower Shortage Areas for primary care. Other shortages exist in 
some areas for dental, vision care and psychiatric manpower. 

Fourteen of twenty urban National Health Service Corps sites within the state 
in 1980 are located in the two counties. 

General Hospital Utilization has differed significantly between these two 
counties and the rest of the state since 1970. 

From 1970 to 1982 Two Counties Rest of State 

% Change in ambulatory visits +5 +75 

% Change in patient days -18 +18 

This suggests marked reduction in use of general hospitals, even while the 
physician population grew 7.4 percent from 1960 to 1970, and another 
7.7 percent from 1970 to 1980 while the total population fell by 5.8 percent. 
The increase in physician population from 1960-1980 does not recover the 
decrease of 26.7 percent noted between 1950 and 1960. 

427 



The increase in physician manpower in the two counties was relatively 
insignificant when compared to the increases seen in the rest of the state 
during the same time periods. In other urban counties, the increases were 52 
percent for the years 1960 to 1970 and 19.6 percent for 1970 to 1980, while 
the overall population fell 3.2 percent. This was in addition to the 72 
percent increase seen between 1950 and 1960. 

In 1950, nearly two-thirds of all the state's physicians (18,975) were in 
these two counties. By 1980, even with the aforementioned increases of the 
past two decades, there was an actual decline in the number of physicians 
(16,083) to account for only one-third of the state's total. Indeed, in 1940 
there were more physicians in these two counties (16,889) than in 1980. 

Some measure of the distribution of these changes among physicians can be seen 
by reviewing the numbers of specialists between 1975 and 1979. For example, 
while the number of hospital-based general practitioners in the rest of the 
state increased by 26.1 percent, in these two counties there was a 14.1 
percent decrease observed. Obstetricians/Gynecologists showed a similar, but 
less severe pattern. The numbers of Internists and Pediatricians in 
office-based practice increased modestly (14 percent and 4.7 percent, 
respectively). In the rest of the state, however, there was minimal change in 
these counties (2.9 percent and 1.0 percent, respectively). 

When reviewing all active non- Federal physicians, the total number of 
physicians per 100,000 population in the two counties (471) far exceeds that 
for the rest of the state (212). However, when reviewing perhaps the most 
critical area of specialty for primary care, i.e. Family Medicine/General 
Practice, there is no appreciable difference (22 vs. 20). 

Among other health professionals, the number per 100,000 population fluctuated 
as follows: 

Professions Two Counties Rest of State 

Pharmacists (1974-1980) -23.8% -13.5% 

Employed RN's (1972-77) + 1.1% +25.0% 

Optometrists (1972-1981) -12.2% - 8.4% 

Dentists +10.1% + 9.5% 

California 

Three-fifths of the state's population can be found in these 11 of the 
47 counties. All 11 counties are urban communities. Seven of the 11 counties 
have a population in excess of one million persons while another three 
counties have between 250,000 and one million persons. The eleventh county 
has in excess of 100,000 persons. The total Asian/Pacific Islander population 
for the state is 2.9 percent compared to 7.2 percent within the 11 counties. 

The personal income per capita is somewhat higher in these 11 counties than 
for the rest of the state ($11,376 vs. $10,297). The Aid to Families with 
Dependent Children per 100,000 total population is higher for the 11 counties 
(6,353 vs. 5,063). There is a slightly higher percentage of the population 
below the poverty level in the 11 counties (11.8 percent vs. 10.5 percent). 

428 



The Infant Mortality Rate for "other" (than nonminority or Black) is the same 
for these counties as it is for the rest of the state, i.e. 6.0. The overall 
mortality per 1,000 population is 8.0 in these counties vs. 7.7 for the rest 
of the State. 

Of the eleven counties, a portion of ten of the eleven counties have been 
designated as both Medically Underserved Areas and primary care Health 
Manpower Shortage Areas. Portions of the eight of the eleven counties have 
dental manpower shortages ; two have vision care and two have psychiatric 
manpower shortages; one entire county and part of another have podiatric 
manpower shortages . 

For the rest of the State, five counties and portions of the 19 of 47 counties 
have been designated Medically Underserved Areas, while two entire counties 
and parts of another 39 counties have been designated primary care health 
manpower shortage areas. Entire counties, totaling 21, have podiatric 
manpower shortages; parts of eleven counties have dental shortages; one entire 
and portions of three counties have psychiatric manpower shortages. 

Thirteen of the twenty urban National Health Service Corps sites within the 
State in 1980 were located within the eleven counties, as were five of the 29 
rural sites. 

Physicians were disproportionately concentrated in the eleven counties, 
totaling 261 per 100,000 population, compared to 186 per 100,000 for the rest 
of the State. The distribution of nurses (both RNs and LPNs), based on 
numbers per 100,000 population and per 100 hospital beds seemed more even. 

The distribution of other health professionals more closely approximated the 
distribution of the general population between the 11 counties and the rest of 
the State. 





11 


Counties 


Rest 


of the 


State 


Total Population 


14, 


,715,000 


8; 


,953,000 




Dentists 




9,408 






5,852 




Podiatrists 




9,408 






5,852 




Optometrists 




1,871 






1,250 




Pharmacists 




8,581 






4,771 





Health Professions training institutions are also disproportionately 
concentrated in the 11 counties, including 5 of the 8 medical schools, 4 of 
the 5 dental schools, 54 of 86 nursing schools, the only podiatry school and 
three pharmacy schools and 1 of 2 optometry schools. 



429 



^ 



P-i -H 



a 



*, -U Q ON 
■y vj O rH 



PLI 



m 






6^ S 



PU 



&4 tH 



^1 



8 



§1 

■^^ 4.) 



•H ij a <— ' 

Cd CO CO ^*N 

ilia 

g cj a -H 

ffl -H 'O JJ 

^„ S CO & 

^S di M O 



•iH CO ,-v 

■d w <^ 



PM 



CO 



<f CM 



<r 00 



in cN 






r-! CD 

r^ CM 

r^ CM 

O "^ 



St <^ 

en CM 

m 

CM 






o go 

CM f^ 
CO CM 



s 



■y "^ 

y O o 



CM rH 



O O 

• • 









r3 

en 



CO "^ 
00 



CM C3> 

r^ CM 



iri en 

-< 00 






s 



o 

U-1 



■g 



I, 



430 



PART II. A. 4. - American Indians/Alaskan Natives' Conimunities and Health 
Professional Resources 

American Indians/Alaskan Natives represent the smallest of the four minority 
groups with total numbers, by 1980 Census count, of approximately 1.4 
million. This represents a 70 percent increase over the 1970 Census, which is 
due to increased numbers of people declaring this identification, rather than 
the result of changes in vital events, i.e. births and deaths. 





1970 

792,730 

34,525 

827,255 


1980 


Difference 

Between 
1970 & 1980 

+571,303 
+ 21,842 
+593,145 


% Difference 

Between 
1970 & 1980 


American Indian 

Eskimo, Aleutian Islander 

Total 


1,364,033 

56,367 

1,420,400 


+72.1 
+63.3 
+71.1 



This minority population is geographically spread throughout the continental 
United States and consists of more than 600 tribes and other sub-groups. 
Alaska is the State with the highest percentage of American Indian 
inhabitants, i.e. 16 percent of the total. California, however, has by far 
the largest number with nearly a quarter of a million inhabitants. 

Despite the relatively small numbers, the heterogeneity of the American Indian 
population is quite striking. Customs and cultures vary widely, as do the 
nature and formality of the relationships between these tribes and 
communities, and local, state and Federal governments. Efforts to address the 
concerns of individual Indian communities require information on the specific 
make-up of the residents of that community. Inevitably sub-county and tribal 
data have to be obtained and analyzed. 

Counties having greater than 20 percent American Indian population are spread 
over 10 states. In part this is a reflection of the location of reservations; 
otherwise it reflects a strong tendency to live as a community in individual 
counties rather than dispersing. 

The State of Alaska was excluded from this analysis for essentially two 
reasons. First, it has no "county" structure and by itself has only a 16 
percent Alaskan Native population. Second, as a state it has a different 
resource base than the counties being used for comparative analyses. 

The State of California, although comprising the largest single state 
population of American Indians, was also not a part of this analysis since 
none of its counties met the 20 percent threshold level. 

The working group agreed that future analyses would have to utilize a 
methodology which assured some examination of the Indian populations of those 
two states. However, it was also agreed to apply the same standards at this 
time as applied to county analyses for the other minority groups. 



431 



In all states with sizable American Inciian populations, those counties with 20 
percent and over Indians have substantially lower physician to population 
ratios and greater percentages of persons living below the poverty level than 
all other counties in the states. The counties with these Indian populations 
also show lower numbers of hospital inpatient days and slightly higher infant 
mortality. (See Table 17). 



Table 17 

Comparison Among States of Key Indicators: Counties with 

20 Percent or More American Indians and All Other Counties 





Active M.D. 


Infant 


% Population 


1 
Number | 




Phyf 


sicians/ 


Mort 


ality 


below 


poverty 


Inpatient | 




lOOOK pop. 
MC AOC 


Rate 


level 


days/1 
MC 


000 pop. 1 


1 State 


MC 


AOC 


MC 


AOC 


AOC 1 


1 Arizona 


69 


189 


13.0 


12.3 


28.5 


12.1 


533 


1,123 1 


1 Montana 


59 


137 


19.1 


11.8 


20.4 


11.4 


1,918 


1,482 1 


1 Nebraska 


97 


149 


31.9 


11.3 


23.9 


10.4 


2,143 


1,674 1 


1 New Mexico 


92 


164 


10.9 


11.7 


27.1 


11.6 


1,610 


1,932 1 


1 North 


















1 Carolina 


72 


154 


18.4 


14.4 


24.8 


14.3 


1,067 


1,244 1 


1 North 


















1 Dakota 


67 


139 


11.8 


12.1 


27.1 


11.6 


1,610 


1,932 1 


' Oklahoma 


42 


132 


10.2 


12.8 


22.8 


13.0 


590 


1,273 1 


1 South 


















1 Dakota 


26 


121 


19.4 


10.0 


39.4 


14.7 


448 


1,915 1 


1 Utah 


49 


170 


8.5 


10.4 


32.9 


10.3 


489 


815 1 


1 Wisconsin 





161 


8.9 


10.3 


NA 


8.5 


NA 


1,399 1 


1 Average 


57 


152 


15.2 


11.7 


27.3 


12.3 


1,053 


1,389 1 



MC = Minority county(ies) 
AOC = All Other Counties 



When 1970-1980 changes in health indicators are examined, there are three 
mortality areas in which counties with 20 percent and over American Indian 
population are worsening: (a) deaths from cancer are increasing in the 
counties with high proportions of Indians at a faster rate than in all other 
counties; (b) deaths from cirrhosis of the liver increased n average of 45.7 
percent in Indian-population counties but dropped in all other counties but 
dropped in all other counties by an average of 15.1 percent; (c) there was an 
increase in almost all Indian populated counties of deaths in the 15-24 age 
group while at the same time, deaths in this age group were falling in all 
other counties. (See Table 18). 

These findings, from Bureau of Health Professions ARF file data, are 
consistent with Indian Health Services records of trends in Indian Health. 



432 



The total U.S. Indian infant mortality rate has been falling and by 1979 was 
14.1 compared to 13.1 for the entire U.S. Alcoholism-related mortality among 
Indians is six times greater than for the U.S. population. And, common causes 
of deaths of young adults — accidents, suicide and homicide — are significantly 
higher in the Indian population than in the total U.S. population. 

Brief narratives for the 10 states included in the analysis follow. They 
relate the socio-economic disadvantage of the inhabitants of these high 
density American Indian counties, as well as the relative scarcity of health 
professions non-Federal resources. This is followed by a brief discussion of 
the Indian Health Service. 



Table 18 

Compairson Among States of Selected Mortality 

Patterns, 1970-1980: Counties with 20 Percent or More American 

Indians and All or Other Counties 





Mortality 


Rate Percent Change 


1970 - 


1980 






Cirrhosis 
MC 


of Liver 
AOC 


Cancer 


15-24 Ag€ 
MC 


> Group 1 


i state 


MC 


AOC 


AOC 1 


1 Arizona 


+88.6 


-22.5 


+26.6 


+21.5 


-38.1 


-16.6 1 


1 Montana 


+103.9 


-12.9 


+19.0 


+7.8 


+18.5 


+0.7 1 


1 Nebraska 


+92.0 


-29.4 


+2.4 


+9.8 


+100.0 


-11.2 1 


1 New Mexico 


-27.2 


-26.4 


+28.9 


+24.7 


+20.2 


-12.5 1 


1 North Dakota 


+98.0 


+8.8 


+23.0 


+8.3 


+32.7 


+9.3 1 


1 Oklahoma 


-23,6 


-21.8 


+26.2 


+3.4 


+15.3 


-1.1 1 


1 South Dakota 


+20.9 


-3.2 


+3.2 


+9.4 


-7.8 


-17.6 1 


1 Utah 


+55.1 


-30.4 


+93.8 


-3.3 


+62.3 


-22.5 1 


1 Wisconsin 


NA 


NA 


NA 


NA 


NA 


NA 1 




+45.7 


-15.1 


+27.1 


+12.2 


+22.6 


-9.7 1 



MC = Minority County(ies) 
AOC = All Other Counties 



New Mexico 

32 total counties 

Counties with 20 percent or more American Indians: 



McKinley 
San Juan 
Sandoval 



65.7 percent 
33.0 percent 
27.2 percent 



Rural Population. The percentage of rural population in minority counties was 
twice that in all other counties in New Mexico, 48.4 to 24.7. 



433 



Income. Personal income by household was almost equal for both minority 
counties and all other counties, $23,361 and $23,372 respectively. However, 
in minority counties the population percentage below the poverty level was 
26.0 vs. 16.3 in all other counties. 

Health Professions Schools. There are 13 health professions schools in New 
Mexico. Of these, only two nursing schools are in minority counties and, only 
23 of 415 nursing graduates are from the two minority counties' schools. 

Health Practitioners. Active M.D. physicians per 100,000 population are 164 
in all other counties and 92 in minority counties. There are also 
disproportionately small numbers of dentists, veterinarians, optometrists, and 
pharmacists, and there are no podiatrists in minority counties. 

The RN to population ratios are relatively close. 

From 1970 to 1980, the minority counties have shown a greater increase in M.D. 
physicians than in all other counties. The number of minority counties M.D.'s 
increased by 133.3 percent compared to 45.2 percent in all other counties. 
None of the minority counties had a decreased physician/population ratio while 
6 of 29 all other counties had a lower ratio in 1980 than in 1970. The 
minority counties increase held for all M.D. specialities, for example, 
pediatricians increased from 3 to 8 and surgical specialists increased from 15 
to 26 between 1975 and 1979. 

Hospital Utilization. In minority counties ambulatory visits per 1,000 
population were twice those in all other counties but inpatient days were only 
two-thirds of what they were in all other counties. The change in both these 
ratios between 1970 and 1982 widened the difference in both types of 
utilization between minority counties and all other counties in the State. 

Infant Mortality. Infant deaths dropped in both minority counties and all 
other counties between 1970 and 1980 but are slightly higher in all other 
counties, 11.7 to 10.9 per 1,000 live births. 

Utah 

29 total counties 

County with 20 percent or more American Indians: 

San Juan 45.9 percent 

Total Population. The one minority county in Utah contains only 12,000 
persons of whom 5,500 are American Indians. 

Income. 32.9 percent of the minority county population lives below the 
poverty level compared to 10.3 percent of all other tween 1970 and 1980 but 
are slightly higher in all other counties, 11.7 to 10.9 per 1,000 live births. 

Key County. Personal income by household is about $6,000 less. Thirteen 
percent of the minority county population receives AFDC while only two percent 
of all the other counties population are recipients. 

434 



Health Practitioners. Although the minority county has only a 12,000 
population one would expect more health practitioners than there are: six 
general practitioners and no surgeon, osteopath, dentist, veterinarian, 
podiatrist, optometrist or pharmacist. All other counties appear to have a 
normal complement of those practitioners. There are nurses in the minority 
county but they too are well below the ratio for the rest of the State. 

Hospital Utilization. Minority county inpatient days per 1,000 population are 
low and dropped almost 50 percent between 1970 and 1982. In all other 
counties inpatient days increased. 

Infant Mortality. Between 1970 and 1980 American Indian infant mortality fell 
from 31.6 to 6.1. This 1980 rate is below the total for all other counties 
and the rate for nonminorities and Blacks. 

Other Mortality. There were increases between 1970 and 1980 in minority 
county death rates from cirrhosis of the liver and cancer which did not occur 
in all other counties. There were large increases in the death rates for 
several age groups but this may be attributable to the small size of the 
population. 

Wisconsin 

71 total counties 

County with 20 percent or more American Indians: 

Menominee 89.4 percent 

Rural Population. The counties minority population is 100.0 percent rural in 
a State which is 35.8 percent rural. 

Total Population. About 60 percent of the 3,000 people in the minority county 
are younger than 25. In all other counties only 42 percent of the population 
is under 25. 

Income. The AFDC level in the minority county is extremely high: 31 percent 
of the population are receiving assistance compared to 4 percent in all other 
counties. However, those below the poverty level in the minority county 
dropped by 38 percent between 1969 and 1979 vs. a 6 percent decrease in all 
the other counties. 

Health Practitioners. In a minority county of 3,000 people the only health 
practitioner is one general practitioner. There are no others, not even a 
nurse. 

Hospital Utilization. Data are not available. 

Infant Mortality of American Indian in the minority county has decreased by 
half, to 9.3. Total infant mortality for the rest of the State is 10.3. 

Other Mortality. Data are not available. 



435 



Arizona 

14 total counties 

Counties with 20 percent or more American Indians: 

Apache 74.9 percent 
Navajo 47.6 percent 
Coconino 27.9 percent 

Rural Population. Although the State population is largely urban, 57 percent 
of the counties minority population is rural. 

Income. There is a rather wide disparity between minority counties and all 
other counties poverty population percentages. 28.5 percent of the population 
in minority counties and 12.1 percent in all other counties are poor. This 
population increased in both areas though by only 19 percent in minority 
counties and 35 percent in the rest of Arizona. 

Health Professions Schools. In the minority counties there are two nursing 
schools graduating 36 nurses in 1979. There are also a number of allied 
health schools. 

Health Practitioners. Arizona's non-twenty percent or more American Indian 
population counties have the highest M.D. ratio, (189), of all States 
considered in this analysis but the minority counties M.D. ratio is still only 
69. There are no podiatrists but sizable numbers of dentists, optometrists 
and pharmacists. The ratio of nurses approximates that of minority counties 
in other States which have nursing schools located in them. 

In the late 1970s, there were increases in all health professionals except 
pharmacists. 

Hospital Utilization. In the minority counties, the ratio of ambulatory 
visits to inpatient days is six to one. In all other counties in Arizona the 
ratio is closer to one and one-half to one. 

Infant Mortality. In both minority counties and all other counties infant 
mortality is about equal but it appears that American Indian infant mortality 
in the minority counties has been reduced more than that for any other group. 

Other Mortality. Cirrhosis of the liver mortality dropped by 23 percent in 
all other counties but increased by 89 percent in minority counties. 

North Dakota 

53 total counties 

Counties with 20 percent or more American Indians: 

Sioux 64.7 percent 

Rolette 57.6 percent 
Benson 28.7 percent 

Rural Population. The entire counties' minority population is rural. 
One-half of all the other counties population is rural. 

436 



Income. Almost 14 percent of minority counties population receives AFDC vs. 
one and one-half percent of all other counties population. The minority 
counties poverty population is 27 percent compared to 11.6 percent in all 
other counties. In both areas the poverty population diminished by about the 
same amount between 1970 and 1980. 

Health Practitioners. The M.D. physician population ratio in all other 
counties is twice that in the three minority counties, 139 to 67. Minority 
counties M.D.'s did increase by 8, 125 percent, between 1970 and 1980. The 
only M.D.'s identified in the minority counties are GP's; there are no 
specialists in a population of 24,000. 

Hospital Utilization. There is an extremely high ambulatory visit ratio in 
the minority counties — seven and one-half per year for each resident. This is 
so even though the inpatient day ratio is nearly as high as in all other 
counties. 

Infant Mortality. For both the minority counties and all other counties the 
rate is about the same, 11.8 and 12.1. 

Other Mortality. In the minority counties death from cirrhosis of the liver 
rose by 98 percent between 1970 and 1980 and there wee marked increases in 
death rates for certain age groups, including the 15-24 ages. 

Nebraska 

93 total counties 

County with 20 percent or more American Indians: 

Thurston 34.0 percent 

Rural Population. The minority county rural population is 100 percent rural; 
in all other counties the rural population is 36.8 percent. 

Income. Nine percent of the minority county population receives AFDC 
assistance while only two percent of all the other counties population 
receives AFDC. 

The personal income by household in the minority county is only about $2,500 
lower than that for all other counties. The poverty level difference, 
however, is between 23.9 percent (minority county) and 10.4 percent (all other 
counties) . 

Health Practitioners. With a total population of 7,000 the minority county 
has no dentists, veterinarians, podiatrists, or optometrists. The active M.D. 
physician population ratio is 97 per 100,000, highest among the minority 
county considered in this analysis. 

Hospital Utilization. Both in ambulatory visits and inpatient days the 
minority county population is well above the rest of the State, e.g., the 
ambulatory visit ratio is three and one-half times that of all other counties. 

437 



Infant Mortality. Even though hospital utilization is very high in the 
minority county, its infant mortality rate is 31.9 compared to 11.3 in all 
other counties. 

Other Mortality. Cirrhosis of the live was 92 percent higher in 1980 than in 
1970 in the minority county. It decreased by 30 percent in all other 
counties. The minority county 15-24 and 35-44 age groups showed marked 
increases in 1980, 100 percent and 167 percent, respectively. 

Oklahoma 

75 total counties 

Counties with 20 percent or more American Indians: 



Adair 

Cherokee 

Delaware 



33.4 percent 
26.0 percent 
20.7 percent 



Rural Population. 82.2 percent of the minority counties population is rural 
vs. 31.5 percent of the population in all other counties. 

Total Population. The minority counties and all othe counties populations are 
similar in that both are predominantly nonminority. The minority counties 
population is 26.2 percent American Indian; all the other counties it is 5.2 
percent American Indian. 

Income. Per capita income and personal household ncome in the minority 
counties are substantially below that in all other counties. In fact, 
household income in all other counties is over $10,000 more than in the 
minority counties. 

Between 1969 and 1979, the poverty level populations of both minority counties 
and all other counties were diminished by about 15 percent but there is still 
a larger percentage of the minority counties population defined as living in 
poverty — 22.8 to 13.0 percent. 

Health Professions Schools. As in most other American Indian minority 
counties, there are no health professions schools. 

Health Practitioners. The number of minority counties M.D.'s had remained 
relatively stable between 1940 and 1970 but between 1970 and 1980 increased by 
12, or 75 percent. There were also increases in minority counties 
pharmacists, optometrists, dentists, and veterinarians. 

Hospital Utilization. The inpatient days to ambulatory visit ratio is about 
one to one in all other counties but one to three in minority counties. 

Infant Mortality. Infant mortality is slightly lower in the minority counties 
than in the rest of Oklahoma. 

Total Mortality. Deaths of young adults and deaths by motor vehicle accidents 
were significantly increased between 1970 and 1980 in minority counties but 
not in all other counties. 



438 



Montana 

57 total counties 

Counties with 20 percent or more American Indians: 

Big Horn 46.2 percent 

Glacier 45.9 percent 

Roosevelt 36.9 percent 

Blaine 31.8 percent 

Rosebud 24.3 percent 

Rural Population. A large portion of the population in all other counties is 
rural (45.3 percent) but an even larger part of the minority counties 
population is rural (74.5 percent). 

Income. Per capita income in both minority counties and all other counties 
are about the same and have increased from approximately the same base in 
1970. However, the percentage of population in poverty is almost twice as 
high in the minority counties (20.4) as in all other counties (11.4). 

Health Professions Schools. There are none in the five minority counties and 
there are no junior colleges or colleges. Montana has 13 colleges, junior 
colleges, and universities, four RN nursing schools, one pharmacy school and 
numbers of allied health schools. 

Health Practitioners. Considering the 49,000 population of the minority 
counties, there appear to be adequate numbers of dentists, optometrists, 
pharmacists, and nurses. The M.D. physician population ratio, however, is 49 
compared to 137 in all other counties. 

Hospital Utilization. The pattern in ambulatory visits and inpatient days is 
similar to that between most other American Indian minority counties and all 
other counties: the minority counties population is much higher in ambulatory 
visits and lower in inpatient days. 

Infant Mortality. Infant mortality among the minority counties American 
Indian population remains much higher than it is for the total of all other 
counties population — 18.4 to 11.8, even though it was reduced by one-half 
between 1970 and 1980. 

Other Mortality. Deaths attributable to cirrhosis of the liver increased by 
more than 100 percent while decreasing in all othe counties. The death rate 
for the 15-24 age group also increased significantly in the minority counties; 
it did not in all other counties. 



439 



South Dakota 

67 total counties 

Counties with 20 percent or more American Indians: 



Shannon 

Todd 

Buffalo 

Ziebach 

Dewey 



93.4 percent 
77.6 percent 

70.8 percent 
58.1 percent 

57.9 percent 



Corson 

Jackson 

Mellette 

Bennett 

Lyman 



47. 3 percent 

43.4 percent 
38.8 percent 
38.6 percent 

23.5 percent 



Rural Population. Almost the entire minority counties population is rural 
(93.2 percent) while 50.7 percent of all the other counties population lives 
in rural areas. 

Total Population. Two-thirds of the minority counties population is American 
Indian but only 2.6 percent of the population in othe counties in South Dakota 
is American Indian. 

Income. The poverty population in minority counties is almost 40 percent 
compared to 15 percent in all other counties. The AFDC ratio in minority 
counties is six times higher than in all other counties — 12,814 to 2,292. The 
poverty population increased in minority counties between 1970 and 1980 while 
it was dropping in all other counties. 

Health Professions Schools. There are no health professions schools in the 
minority counties and 43 in the State, including 25 dental assistant schools. 

Health Practitioners. There are no podiatrists or optometrists in the 
minority counties even though the minority counties have a population of 
46,000. There are three dentists and three pharmacists but this is 
substantially below their representation in the rest of the State. 

Even the nurse ratio to population is only a third in the minority counties of 
what it is in all other counties. 

The M.D. physician to population ratio (26) in the minority counties is the 
second lowest among the 8 States being analyzed. 

Hospital Utilization. Ambulatory visits per 1,000 population are 4,157 
compared to 448 inpatient days per 1,000 population in the minority counties. 
Comparable numbers in all othe counties are 1,059 and 1,915. 

Infant Mortality. Infant mortality is twice as high in the minority counties 
as in all other counties. 

Other Mortality. Deaths from cirrhosis of the liver significantly increased 
from 1970 to 1980 and deaths in the 5-14 age group also were much higher in 
1980 than in 1970 in minority counties but not in all other counties in South 
Dakota. 



440 



North Carolina 

100 total counties 

Counties with 20 percent or more American Indians: 

Robeson 34.9 percent 
Swain 24.2 percent 

Rural Population. Three-quarters of the minority counties population is rural 
compared to slightly over one-half all other counties population. 

Total Population. 34.0 percent of theminority counties population is American 
Indian but only 0.5 percent of the population in all other counties in North 
Carolina is American Indian. Almost the entire American Indian population of 
North Carolina is in two minority counties. 

Income. Personal income by household is appreciably higher in all other 
counties, $22,455 to $17,588 although percentage increase since 1970 has been 
higher in the minority counties. Ten percent more of the minority counties 
population than of all the othe counties population is still below the poverty 
level. 

Health Professions Schools. There are no health professions graduates in the 
minority counties. There are also no LPN, dental assistant, or dental hygiene 
schools in the minority counties but 58 in the State. 

Health Practitioners. The active M.D. ratio is twice as high in all other 
counties as in minority counties, 154 per 100,000 population to 72. Both 
minority counties are considered medically underserved and 71 of 98 all other 
counties are MUA's. 

There are 27 National Health Service Corps sites in North Carolina; one in 
minority counties. 

Nurses, like M.D.'s are twice as well represented in all other counties as in 
minority counties. 

In all other counties there has been a gain in all medical specialities but in 
the minority counties there has been a drop in specialists in internal 
medicine, obstetrics/gynecology, and surgery. 

Hospital Utilization. Ambulatory visits are about the same for both minority 
counties and all other counties but inpatient days are roughly 20 percent 
higher in all other counties. 

Infant Mortality. The infant mortality rate in minority counties is high, 
18.4, but has decreased from 25.6 in 1970. The rate in all other counties was 
almost as high in 1970 but fell further by 1980, to 14.4. 

Other Mortality. Mortality for all age groups fell markedly from 1970 to 1980 
in minority counties except the 15-24 age group which rose slightly. 

441 



Indian Health Service 

Another major factor which must be considered in any review of Indian 
communities is the varying presence and role of the Indian Health Service. 
More time and specific data are required to complete the review of the role of 
this important resource. 

The following tables present available information on the location and 
distribution of Indian Health Service resources. 



Table 19 
Physician (M.D.) Per 100,000 Population Ratio and Indian Health 
Service Hospitals and Health Centers, in Counties 
with 20 percent or more American Indian Population 





Non-Federal 


Physician 










Ratio/Counties 






IHS Fac 


ilities i 




20% or more 


Less 


than 20% 


Hospitals 






American Indian 


American Indian 


and 


Med. 


Health | 


i state 


Population 
69 per 100,000 


Population 


Centers 
4 


Centers | 


1 Arizona 




189 


2 1 


1 Montana 


59 per 100,000 




137 




3 


2 1 


1 Nebraska 


97 per 100,000 




149 




1 


1 1 


1 New Mexico 


92 per 100,000 




164 




3 


1 1 


1 North Carolina 


72 per 100,000 




154 







1 


1 North Dakota 


67 per 100,000 




139 




2 


1 


1 Oklahoma 


42 per 100,000 




132 




1 


1 1 


1 South Dakota 


26 per 100,000 




121 




3 


2 1 


1 Utah 


49 per 100,000 




170 







1 


1 Wisconsin 


per 100,000 




161 







1 



In some states with counties having 20 percent or higher American Indian 
population, IHS health services in the counties help to compensate for low 
non-Federal physician ratios. In others like North Carolina, Utah, and 
Wisconsin there are no IHS facilities in those counties with high American 
Indian populations. 



442 



Table 20 

Number of Counties Having Over 10 Percent American Indians 

With Indian Health Service Facilities 



Total Counties 



Hospitals 



Counties With Facilities 



Health Centers 



Either Hospital or 
Health Center 



No. 



No. 



% 



No. 



% 



70 



22 



31,4 



26 



37.1 



44 



62.9 



Although only about a third of the counties have hospitals and a third have 
health centers, there is little duplication of the two. Almost two-thirds of 
the counties have one or the other. Hospitals are much more likely to be 
found in the counties with over 20 percent American Indian population while 
health centers are more common in those counties with under 20 percent 
American Indian populations have neither an IHS hospital nor a health center. 
There are substantial IHS resources available but there is not uniform 
availability to the entire U.S. American Indian population. 

If Oklahoma is excluded from the count because it has no reservations exactly 
two-thirds of the counties have reservations within or on their borders. Of 
19 counties with over 35 percent American Indian population, only one is not a 
reservation county. Only eight of 37 counties with 10-20 percent American 
Indian populations are counties with reservations. 



443 



Table 21 

Counties with Over 10 Percent 

American Indian Population 



American (%) American 
Indian Indian 
Population Population 



County/Stat 


e 
OK 


1980 


1980 


McCurtain 


3,626 


10.0 


Okanogan 


WA 


3,111 


10.2 


Mineral 


NV 


645 


10.4 


Atoka 


OK 


1,345 


10.6 


Neshoba 


MS 


2,515 


10.6 


Pushmataha 


OK 


1,253 


10.6 


Sawyer 


WI 


1,385 


10.8 


Johnston 


OK 


1,171 


11.3 


Rio Arriba 


NM 


3,379 


11.5 


Fremont 


WY 


4,501 


11.5 


Coal 


OK 


719 


11.9 


Graham 


AZ 


2,730 


11.9 


Mountrail 


ND 


917 


11.9 


Osage 


OK 


4,727 


12.0 


Mcintosh 


OK 


1,891 


12.2 


Craig 


OK 


1,834 


12.2 


Hoke 


NC 


2,557 


12.5 


Beltrami 


MN 


3,920 


12.7 


Hill 


MX 


2,293 


12.7 


McKenzie 


ND 


911 


12.8 


Ottawa 


OK 


4,203 


12.8 


Seminole 


OK 


3,719 


13.5 


Hughes 


OK 


1,944 


13.6 


Mayes 


OK 


4,389 


13.6 


Gila 


AZ 


5,154 


13.9 


Latimer 


OK 


1,385 


14.1 


Valencia 


NM 


8,636 


14.1 


Sequoyah 


OK 


4,462 


14.6 


Okfuskee 


OK 


1,627 


14.6 


Alpine 


CA 


169 


15.4 


Alaska 




64,357 


16.0 


Lake 


MT 


3,162 


16.6 


Ferry 


WA 


978 


16.8 


Jefferson 


OK 


2,030 


17.5 


Charles Mix 


SD 


1,709 


17.7 


Caddo 


OK 


5,525 


17.9 


Mahnoman 


MN 


1,008 


18.4 


Roberts 


SD 


2,110 


19.3 


Delaware 


OK 


4,960 


20.7 


Lyman 


SD 


907 


23.5 


Swain 


NC 


2,493 


24.2 


Rosebud 


MT 


2,406 


24.3 


Cherokee 


OK 


7,987 


26.0 



Indian 
Health Service 
Facilities 
Hospital Health Center 



Reservation 

In or Touching 

County 



X 
X 



X 
X 

X (6) 



X 
X 



X 
X 
X 
X (2) 



X 
X 
X 



X 
X 



X 
X 

X (4) 
X 

X 

X 



X 
X 



X 
X 
X 
X 



X 
X 
X 



444 



Table 21 (continued) 







American 


(%) American 






Indian 


Indian 






Population 


Population 


County/Stat 


e 
NM 


1980 


1980 


Sandoval 


9,471 


27.2 


Coconino 


AZ 


20,949 


27.9 


Benson 


m 


2,277 


28.7 


Blaine 


MT 


2,223 


31.8 


San Juan 


NM 


26,866 


33.0 


Adair 


OK 


6,210 


33.4 


Thurston 


NE 


2,443 


34.0 


Robeson 


NC 


35,507 


34.9 


Roosevelt 


MT 


3,865 


36.9 


Bennett 


SD 


1,174 


38.6 


Mellette 


SD 


872 


38.8 


Jackson 


SD 


1,491 


43.3 


San Juan 


UT 


5,622 


45.9 


Glacier 


MT 


4,882 


45.9 


Big Horn 


MT 


5,126 


46.2 


Corson 


SD 


2,459 


47.3 


Nava j o 


AZ 


32,215 


47.6 


Rolette 


ND 


7,020 


57.6 


Dewey 


SD 


3,107 


57.9 


Ziebach 


SD 


1,342 


58.1 


Sioux 


ND 


2,341 


64.7 


McKinley 


NM 


37,115 


65.7 


Buffalo 


SD 


1,270 


70.8 


Apache 


AZ 


39,042 


74.9 


Todd 


SD 


5,688 


77.6 


Menominee 


WI 


3,014 


89.4 


Shannon 


SD 


10,575 


93.4 



Indian 
Health Service 
Facilities 
Hospital Health Center 



X 
X 



X 
X 

X (2) 
X 

X 
X 
X (2) 

X (2) 
X 



X 
X 



X 

X (2) 



X 

X 



Reservation 
In or Touching 
County 

X 
X 

X 
X 



X 
X 
X 
X 
X 

X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 
X 



NOTES: Oklahoma does not have reservations. 

Alaska is not divided by counties therefore the entire state is 
included in the table. 



445 



Table 22 



Recenl-expertenced CtvllUn labor force 1/ In Health Occupations and Health Occupations 
tlosters In the United States by Racial/Ethnic Category: April 1, 1980 



All 
Races 



Total 
minority 2/ 



Black 
(not 
Hispanic) 



Hispanic 





Asian/ 




White 


Native 


Pacific 


Other 


(not 


American 


Islander 


Hinorlty 


Hispanic) 


26,810 


163,220 


6,070 


4.301,270 


560 


1,510 


50 


96.100 


50 


1.230 


60 


17.220 


510 


41.920 


890 


357.840 


190 


3,830 


70 


115.880 


60 


430 


30 


32,920 


40 


560 


10 


23,310 





100 





7,340 


30 


580 


20 


20,330 


3,860 


42.950 


1.270 


1,114,310 


250 


6,580 


170 


130.440 


330 


2,480 


60 


47,510 


210 


970 


30 


39,840 


50 


430 


20 


16.160 


110 


930 


10 


37,800 


80 


300 


30 


38,100 


210 


620 


60 


35.650 


210 


630 


30 


24,810 





360 


40 


8,060 


70 


350 


10 


17,520 


860 


13,500 


210 


191,210 


40 


550 


20 


44,130 


180 


440 


30 


12,420 


380 


1,060 


100 


82,100 


2,790 


5,900 


410 


333.130 


890 


3,970 


240 


120.630 


830 


2,410 


80 


139.750 


1,920 


4,830 


250 


215.270 


11,690 


19,700 


1,740 


901,210 


190 


910 


20 


40,660 



Total, Health Occupations 5,412,160 

Managers, Medicine 6 Health... 110,800 

Medical Scientists 20,070 

Physicians 433.260 

Oent Ists 125 ,290 

Veterinarians 31,360 

Op tome Ir Is ts 24 ,6 10 

Podiatrists 7,780 

Health diagnosing 

practitioners n.e.c 21,510 

Registered nurses 3/ 1,205,300 

Pharmac Ists 115,610 

Dietitians and dietetic , 

technicians 67,270 

Inhalation therapists 48,710 

Occupational therapists 17,760 

Physical therapists 13,000 

Speech therapists 41,300 

Therapists n.e.c 43,070 

Physicians' assistants 4/ 30,410 

Hedlcal science teachers 9,010 

Health specialties teachers... 19,550 
Clinical laboratory 

technologists and 

techn Ic lans 213,980 

Dental hyglenlsts 16,190 

Health record technologists 

and technicians 15,150 

Radiologic technicians 96,310 

Licensed practical nurses 435,100 

Health technologists and 

technicians n.e.c 152,510 

Dental assistants 153,120 

Health aides, except nursing.. 292,050 
hurslng allies, orderlies 

and attendants 1,370,120 

Optical goods workers 16,870 

Dental lab and medical 

appliance technicians 10,750 



1,140,900 

11,800 
2,050 

75,110 

9,110 

1.130 

1,290 

110 

1,210 

170,990 

15,210 

19,750 
8,910 
1,600 
5,290 
3.210 
7,110 
5,630 
930 
2.030 



52,770 
2,060 

2,730 

11.210 

102,010 

31,910 
18.380 
76.780 

176,920 
6,220 

9,000 



73,503 

9,180 

880 

13,240 

3.130 
520 
250 
280 

260 

95,370 

1,720 

11,100 
5,110 
770 
2.930 
2,010 
5,100 
3,010 
260 
1.230 



28,000 
700 

1,180 

7.900 

77,050 

20,170 

6,640 

52,300 

372,210 
1.960 

2,770 



209.710 

3.100 

630 

18,850 

2,190 

390 

430 

60 

320 

27,510 

3,190 

2,100 
2,590 

330 
1,280 

710 
1,390 
1.750 

280 

370 



10,100 
750 

600 

4,000 

15,060 

6,610 

8,390 

17,180 

71,500 
3,110 

3.760 



220 



2,280 



50 



39.670 



1/ The "recent experienced" civilian labor force Is defined as civilian persons employed In 1980 or uneir^loyed having civilian 
work experience between 1975-1980. 

2/ Includes all race/ethnic Ity categories other than white. 
Due to Independent rounding figures may not add to totals. 

3/ According to data from the 1900 National Sanple Survey of Registered Nurses the total nunfcer of R.N.'s enip'oyed In nursing In 
1900 was 1,272,851. Of these 1,151,221 were white (non-hlspantc), 51,585 Black (non-hlspanic) 30,470 Aslan/Paclfic Islander. 
3.015 American Indian/Alaskan Native, 17,938 Hispanic and 15,592 of unknown race/ethnlclty. 

4/ According to data published In the Third R epo rt to the P resident and Con gr ess on the Status of Health Professions Personnel 
there were 11,000 PAs In the U.S. of whom MOO were esDmated to be active. 



446 



Table 23 



Percent Distribution of the Recent-experienced Civilian labor Force 1/ In Health Occupations and Health 
Occupations Clusters In the United States by Raclsl/Ethnic Category: April I, 1980 







Black 






Asian/ 




White 


All 


Total 


(not 




Native 


Pacific 


Other 


(not 


Races 


minority 2/ 


Hispanic) 


Hispanic 


American 


Islander 


Minority 


Hispanic) 



21.0 



13.5 



3.9 



0.5 



3.0 



0.1 



79.0 



Total, Health Occupations 100.0 

Managers, Medicine (. Health... 100.0 

Medical Scientists 100.0 

Physicians 100.0 

Dentists 100.0 

Voter Inar lans 100.0 

Optometrists 100.0 

Pod latr Is ts 100.0 

Health diagnosing 

practitioners n.e.c 100.0 

Registered nurses 3/ 100.0 

Pharmac Is ts 100.0 

Dietitians and dietetic 

techn Ic lans 100.0 

Inhalation therapists 100.0 

Occupational therapists 100.0 

Physical therapists 100.0 

Speech therapists 100.0 

Therapists n.e.c 100.0 

Physicians' assistants 4/ 100.0 

Medical science teachers 100.0 

Health specialties teachers... 100.0 
Clinical laboratory 

technologists and 

techn Ic lans 100.0 

Dental hyglenlsts 100. 

Health record technologists 

and teclinlclans 100.0 

Radiologic technicians 100.0 

Licensed practical nurses 100.0 

Health technologists and 

technicians n.e.c 100.0 

Dental assistants 100.0 

Health aides, except nursing.. 100.0 
Nursing aides, orderlies 

and attendants 100.0 

Optical goods workers 100.0 

Dental lab and medical 

appliance technicians 100.0 

1/ The "recent experienced" civilian labor force Is defined. as civilian persons employed In 1980 or unemployed having 
civilian work experience between 1975-1900. 

II Includes all race/ethnic Ity categories other than white. 
Due to Independent rounding figures may not add to totals. 

3/ According to data from the 1980 National Sample Survey of Registered Nurses the distribution of enployed R.N. 'S by 
" ractal/ethnic categories was: Wl.lte (non-hlspanic) 90.4 percent. Black (non-hlspanic) 4.3 percent, Aslan/Pacif Ic 

Islander, ^.^ percent. American Indian/Alaskan Native, 0.2 percent, Hispanic 1.4 percent, and 1.2 percent of 

unknown rarlal/ethnlc background. 

4/ According to data published In the Third Reportto the Pr esident and Congress on the Status of Health Professions 
Personnel there were 11,000 PAs In Uw ^.T. oH wfioir8,BiXy were eifUaled lote active. 



13.3 


8.5 


2.9 


0.6 


1.4 


0.0 


86.7 


14.2 


4.4 


3.1 


0.2 


6.1 


0.3 


85.8 


17.4 


3.1 


4.4 


0.1 


9.7 


0.2 


82.6 


7.5 


2.5 


1.7 


0.2 


3.1 


0.1 


92.5 


4.2 


1.5 


l.l 


0.2 


1.3 


0.1 


95.8 


5.2 


1.0 


1.7 


0.2 


2.3 


0.0 


94.7 


5.7 


3.6 


0.8 


- 


1.3 


- 


94.3 


5.6 


1.2 


1.5 


0.1 


2.7 


O.l 


94.4 


13.3 


7.4 


2.1 


0.3 


3.3 


0.1 


86.7 


10.4 


3.2 


2.4 


0.2 


4.5 


0.1 


B9.6 


29.4 


21.4 


3.7 


0.5 


3.7 


0.1 


70.6 


18.3 


10.5 


5.3 


0.4 


2.0 


0.1 


81.7 


9.0 


4.3 


1.9 


0.3 


2.4 


0.1 


91.0 


12.3 


6.8 


3.0 


0.3 


2.2 


0.1 


87.7 


7.8 


4.9 


1.7 


0.2 


0.9 


O.I 


92.3 


17.2 


11.8 


3.2 


0.6 


1.4 


0.1 


82.8 


18.5 


9.9 


5.7 


0.7 


2.1 


0.1 


81.5 


10.3 


2.9 


3.1 





4.0 


0.4 


89.7 


10.4 


6.3 


1.9 


0.4 


1.8 


0.1 


89.6 


21.6 


11.5 


4.1 


0.4 


5.5 


0.1 


78. 4 


4.5 


1.5 


1.6 


0.1 


1.2 


0.0 


95.5 


18.0 


9.8 


4.0 


1.2 


2.9 


0.2 


82.0 


14.8 


8.2 


4.2 


0.4 


1.9 


O.I 


85.2 


23.4 


17.9 


3.5 


0.6 


1.4 


0.1 


76.6 


20.9 


13.2 


4.4 


0.6 


2.6 


0.2 


79.1 


11. 6 


4.2 


5.3 


0.5 


1.5 


0.1 


88.4 


26.3 


17.9 


6.0 


0.7 


1.7 


0.1 


73.7 


34.6 


27.0 


5.2 


0.8 


1.4 


O.I 


65.4 


13.3 


4.2 


6.7 


0.4 


1.9 


0.0 


86.8 


18.6 


5.7 


7.7 


0.5 


4.7 


0.1 


81.4 



447 



PART II. B. - Minority Health Professionals 

One factor which the working group has viewed to be crucial in assuring the 
availability (and enhancing accessibility) of health professionals to minority 
populations and communities, is the availability of minority health 
professionals who reflect the racial, ethnic, cultural and other 
characteristics of those communities. Later in this paper a discussion Is 
presented to support this view. 

At this point, however, an effort was made to briefly present some statistics 
reflecting the degree of the participation of these minority groups in the 
health professions. Since the actual presence of these health providers in 
minority communities was the desired measure of availability, data were 
obtained to document the degree of this presence. Part II. B. 1 presents this 
discussion of Distribution of Minority Health Professionals, while Parts 
II. B. 2 and II. B. 3. respectively, discuss the Development of these 
professionals and their Practice Patterns. 

PART II. B. 1. - Distribution of Minority Health Professionals 

A review of data from the 1980 Census shows, from a national perspective, 
varying degrees of participation among the four minority groups, among the 
several health professions and occupations listed. Tables 22 and 23 provide 
the detail of these census data, by number of persons and percent 
distribution, respectively. 

Among mental health professionals, there are only small percentages of 
minorities. Asian-Americans are represented in psychiatry and almost six 
percent of social workers are Black, but otherwise mental health practitioners 
are almost all nonminority. See Table 24. Both at the M.A. and Ph.D. levels 
there are minimal numbers and percentages of minority clinical psychologists. 
See Table 25. The overwhelming majority of staff in these units are 
nonminority. There are slightly more Native Americans and Hispanics in the 
alcoholism only units than in the combined units. See Table 26. 

With few exceptions, the census data show underrepresentation of the minority 
groups when compared to their percentages in the total population. This is 
especially so for the more prominent professions, i.e. among registered 
nurses, physicians, dentists, etc. The degrees of underrepresentation vary 
widely, but the generally common thread is that the minority numbers are 
disproportionately low. 



448 



Table 24 
Percent Practicing Professionals by Race/Ethnicity 











Asian/ 








Non- 




American 


Pacific 








Minority 


Black 


Indian 


Islander 


Hispanic 


Unknown | 


1 Psychologists 


95.6 


1.4 


0.2 


1.1 


0.7 


0.9 1 


1 Psychiatrists 


69.4 


1.5 


0.4 


5.6 


2.5 


20.7 1 


1 Social Workers 


88.5 


5.8 


NA 


1.6 


1.8 


2.2 1 


1 Nurses 


91.5 


3.7 


0.3 


2.0 


1.2 


1.3 1 



NOTE: Primary data sources are membership surveys of the American 

Psychological Association (1982), American Psychiatric Association 
(1977), National Association of Social Workers (1982), a HRSA survey 
of registered nurses (1980), and an NLN survey of 1980-81 graduations 
of minority students from basic baccalaureate nursing programs. 



Table 25 
Number of U.S. Clinical Psychologists 











Asian/ 








Non- 


American 


Pacific 




1 Level 


Total 


Minority Black Hispanic 


Indian 


Islander 


Unknovm | 


1 Ph.D. 


16,519 


15,885 159 164 


182 


6 


123 1 


1 Master's 


2,486 


2,360 38 19 


29 


8 


33 1 






Percent U.S. Clinical Psychologists 














Asian/ 








Non- 


American 


Pacific 




1 Level 


Total 


Minority Black Hispanic 


Indian 


Islander 


Unknown | 


1 Ph.D. 


100.0 


96.2 1.0 1.0 


1.1 


0.4 


0.7 1 


1 Master's 


100.0 


94.9 1.5 0.8 


1.2 


0.3 


1.3 1 



449 



Table 26 

Racial and Ethnic Characteristics of Staff in Units Providing 

Alcoholism Treatment Only and in Units Providing 

Combined Alcoholism and Drug Abuse Treatment 





Alcoholism Only 
No. % 


Combined 


Total i 


1 Race/Ethnicity 


No. % 


No . % 1 


1 American Indian 


718 2.5 


252 1.5 


970 2.2 1 


1 Asian/Pacific Isl 


308 1.1 


193 1.2 


501 1.1 1 


1 Black 


3,719 13.3 


2,116 13.3 


5,835 13.3 1 


1 Hispanic 


1,414 5.0 


593 3.7 


2,007 4.5 1 


1 Nonminority 


21,608 77.8 


12,721 80.1 


34,329 78.6 | 


1 Total 


27,767 100.0 


15,875 100.0 


43,642 100.0 1 



The exception to the aforementioned trend appears to be the Asian/Pacific 
Islander minority group which seems frequently "over represented." One 
concern in making such a statement, however, is that closer review of selected 
census files shows significant variation between the subgroups of both 
Asian/Pacific Islander and Hispanic health professions. Table 27 displays 
this data: 



Table 27 

Percent Distribution of Hispanic and Asian/Pacific Islander 

Professionals by Subpopulation, 1980 







Physicians 


Dentists 


Pharmacists | 


1 Hispanic 




100 


100 


100 1 


1 -Mexican 


19 


25 


41 1 


1 -Puerto Rican 




8 


4 


11 1 


1 -Cuban 




26 


28 


23 1 


1 -Other Spanisti 


I 


48 


40 


25 1 


1 Asian/Pacific 


Islander 


100 


100 


100 1 


1 -Japanese 




5 


39 


18 1 


1 -Chinese 




16 


29 


41 1 


1 -Filipino 




25 


8 


7 1 


1 -Korean 




10 


8 


10 1 


1 -Asian Indian 




37 


13 


18 1 


1 -Vietnamese 




2 


1 


3 1 


1 -Hawaiian and 


other 


5 


2 


3 1 


1 Pacific Islander 









Thus apparently while some subgroups appear quite overrepresented, others show 
patterns similar to other underrepresented minorities. 



450 



Again, more detailed data are required to complete the analysis and answer 
specific questions for specific communities of subpopulations. Nonetheless, 
the process was continued to review and display the data which were available. 

The four graphs which follow present national statistics for minority 
physicians (Graph 1), dentists (Graph 2), pharmacists (Graph 3) and registered 
nurses (Graph 4) per 100,000 minority population for all four minority 
groups. [Note with caution - the scale of magnitude (Y-axis) changes with 
each graph. ] 



451 



U.S. Manpower/Population Ratios Physician/100, OOP Ethnic/Racial Population 

In 1975 (1970), there were 731.2 Asian/Pacific Islander physicians per 100,000 
Asian/Pacific Islander population. By 1980 this was 1,197.4. For Black 
Americans this figure per 100,000 Black population was 26.6 in 1970 and 50.7 
in 1980. For American Indians the figure was 22.9 in 1970 per 100,000 
American Indian population and 36.0 in 1980. For Hispanic Americans the 
figure was 113.9 per 100,000 Hispanic population in 1970 and 129.1 per 
Hispanic population in 1980. For nonminorities (which includes 96.26 percent 
nonminority Americans) the figure in 1970 was 146.4 per 100,000 nonminority 
population and in 1980 it was 197.8. 



Graph 1 
Minority Physicians per 100,000 Minority Population 
in the U.S., 1975 and 1980 







1197 


























00 - 


731 


'■i *'■■ '' 




- 




'■••..'■■■■,' 
■■•••..■■■•■■,■'■- 
'•■■..■'•■•■.■■■•■ 
'■■■,'■■■■.."■■•■ 


198 




■-/,./ 




114 


129 




±40 


1. '.. '. 






~~T — T"* 




' ^' .■' 


00 _ 




' ■■. \ ■'. 




■■' ,■'' *■' 


'. ''«. '■. 




/ .■■■'' ^ 


'-. \ ''•! 




•■•''"/..•■ 




51 


,■■•■' .■•■■' .•' 


■i *'■ ' 










' r'' .•'' 


'"'•. '■'. 




'■■ ''■. "' 




.■• y 


\ '■•. '■ 


36 


' / / / 


'"' '*■■ ' 




•■■■/■".'.■••■■' 


'-. '*-. ■"-. 


27 






•■■■'.•■''/' 










•■■■ ■■'■' ,■••■ 


;■,■;•■■•,•• 


- 


23 


\\\ 






//.. 


i'V-v' 






'■ '■'■• ••, 




•■■■■'.■■'"■■■' 


\\-^ 






•■-.V;-.. 




. ■■■' .'■■' .•' 


'•. '■. *• 


10 _ 




\ '■■. '■ 










'"//■ 








'■■•■•..'■■•■■■> 

'.. ■». ■•. 
*■' "'■. 






'■■ '*■ * 




/,/"/ 


'■-. '•• ■'•■ 




,'■■,. ••'■'..•■ 


••■..\. '•■■•., 




f' ,.'■' .y 


'■ ''•■ \ 




.••'''./■•• 


'•■ '■- '*"■. 




.•■•''..■■■'■'■••■■ 


'•- \ '•■. 




y' / .' 


. ' ■• '"'•-. 




..■■■ / . 












.•■••■ / , 


' "i 




••■' •■■■ . 


'• ''l 










i^<i 


>c^;^ 




■'7/ 






^^ 


^ 




'% 












■■'" .•-' .-■' 


•■■..\. ■•., 




'"-'"/ 


••• > '■•■. 




/..'•■..■• 


■■■>s 




/ ^ ■■•' 


•-, 'V./'-y 






' ■■'. '■•. 




*' '" .' 


•.. S. V. 




■-' / ■' 


• \^ 




■■•■'' ,''' •' 


• •: '■■■ 






■ '■■ '■% 


1 _ 


■•/■/ 


'■•■. '■■•.. 

^-—^ — ^ 




>/•/ 


\ '■■■, ■'. 




' / ■•■' . 


, '■■■■■, "V 




' / /\ 


'■■■ '•■i '■ 




/ /\. 


•■. '■■■ *■ 



Asian 



Black Msp 

IZZ] 1975 



Amer. Indian 
■'^■' 1980 



Nanmin. 



452 



U.S. Manpower/Population Ratios Dentist/IQO.OOO Ethnic/Racial Population 

In 1970 there were 89.4 Asian/Pacific Islander dentists per 100,000 
Asian/Pacific Islander population. In 1980 the number was 109.4 Asian/Pacific 
Islander dentists per 100,000 Asian/Pacific Islander population. For Black 
Americans in 1970, there were 10.5 Black dentists per 100,000 Black American 
population. In 1980 this figure was 12.0 per 100,000 Black population. In 
1970 there were 8.3 American Indian dentists per American Indian population. 
In 1970 there were 11.3 Hispanic dentists per 100,000 Hispanic population. 
This figure was 15.0 in 1980. For nonminority dentists the figure was 50.0 
per nonminority dentists per 100,000 nonminority population in 1970. In 1980 
this figure was 64.1 per 100,000 nonminority dentists. 



Graph 2 

Minority Dentists per 100,000 Minority Population 

in the U.S., 1975 and 1980 



120 

UO- 

100- 

90- 
80- 
70 
60- 
50- 
40 
30 
20 
10 H 




109 



89 



•'■■ / / 



• / / 



y// 



/ 



■/A 



./' 



y/. 






Asian 



U 



12 



11 E 



15 



13 



'A 



'^■^i^ 






yy..\ 



\ 



\ \ 



Black 



Hisp. 



Amer. Indian 



64 



50 



/ . • 






/' 






s 
s. ■ ■ 



Ifonmin. 



IZZl 1975 



1580 



453 



U.S. Manpower/Population Ratios Pharmacists/ 100, OOP Ethnic/Racial Population 

In 1975 there were 129.8 Asian/Pacific Islander pharmacists per 100,000 
Asian/Pacific Islander population. In 1980, the number had increased to 
187.9. The Black ratio increased by almost 50 percent between 1975 and 1980. 
In 1980, it was 18.1. Hispanic pharmacists continued to have the second 
highest ratio per 100,000 population among minority groups. Their ratio was 
23.1 in 1975 and 29.1 in 1980. The American Indian ratio increased only 
slightly during the five-year period, 16.6 to 17.6. For nonminority 
pharmacists, the figure was 60.0 per 100,000 nonminority population in 1975 
compared to 72.1 in 1980. 



Graph 3 

Minority Pharmacists per 100,000 Minority Population 

in the U.S., 1975 and 1980 



200 

190 

180 

170. 

160. 

150 

140 

130 

120- 

UO. 

100- 
90- 
80- 
70. 
60. 

50- 
40- 
30- 
20. 
10. 
0. 



188 



130 






72 



60 



rr 



01 


29 




18 




■■•. '■■. '■ 


1-7 1« 


12 






■•■// 








,■•-..■•■•••..■■• 








*■■ ■*•. V 

•. '. — 1 


/// 



Asian 




Amer. Indian 



Ncxmin. 



':J 1980 



454 



, S. Manpower/Population Ratios R.N. 's/100,000 Ethnic/Racial Population 



The Asian/Pacific Islander R.N. ratio increased between 197 
almost 450 nurses per 100,000 Asian/Pacific Islander popula 
ratio was 1,226.8. The Black ratio rose 26 percent to 365 
population. Hispanic nurses had a smaller rise between 197 
175.5 to 188.6. The American Indian R.N. ratio per 100,000 
population is higher than Hispanics though lower than other 
in the table. Their ratio increased from 240.7 to 272.2 
R.N. ratio had the second highest increase (after Asian/Pac 
45 percent to 616.0 per 100,000 nonminority population. 



5 and 1980 by 
tion. In 1980, the 

4 per 100,000 Black 

5 and 1980, from 
American Indian 
groups considered 

The nonminority 
ific Islanders), by 



Graph 4 
Minority R.N.'s per 100,000 Minority Population 
in the U.S., 1975 and 1980 



1.5-1 
1.4- 
1.3- 




1??7 


























1.2- 

1.1" 
1- 

0.9- 

0.8- 

0.7- 


789 


**-. * . 
' ■••. ' '••, 

' "■■• '■-. 
■■. ■■• ' 

•■■.,,■■■••.. N 

'•. ■'. 

""•..'■■■■."' 
"... '•■.. \ 

■•i '■■■ *' 
'■■- '■ "•' 

''\ ■•', 
\'\\ 

\ ■■■. ' 






,-■' /'' 

' ' y ■'' 

//.■'■ 

•■".■''"•■■■ 

/.-•■''..•■' 
. ■' •■■' •■ 
.■'■'.'■ . 

•■' / y' 
.t .-• ^ 

•'' / -■'' 
/ /.'■ 

,'■ y f 
' /..' 




616 


0.6- 
0.5- 




425 


.'••■. \ 
■-. ■'■ ■• 

•W 

■■•■.•■■.' 
■■■■...■•■■•..■■ 


0.4- 




365 






yy 

.■■" •''' -■' 
■■'* .■■* -* 

./' .-■' ,. 

■/.,'. 

• y' y 


0.3- 




289 


'■■.'■■■."■■• 
'■ "^ '■■•. 

..w 

'■■■. \ ' 


241 


272 






/ •■■' 
' / ••■■ 










0.2- 
0.1- 




176 -L^^ 




■ / ' ' 




■ f . • 

■ ,/ ,.••■ 


:^- 




0- 































Asian 



Black Hisp. Amer. Indian 
1975 [33] 1980 



Nomdn. 



labels en bars are actual numbers, not numbers in thousands. 



455 



The trends in minority enrollment in health professions schools are similar to 
those described above for practitioners. It is unlikely that in the near 
future the proportions of practicing minority health professionals will 
increase at a higher rate than they have since 1975. 

Just as the national trends showed increases in professionals for each of 
these disciplines among the nonminority population, so too were there 
increases in each minority group. However since the increases were "across 
the board," the gap between nonminority and minority was not appreciably 
changed . 

As a second phase in reviewing the availability of minority health 
professionals, an effort was made to relate information based on analyses at 
lower than national level. A particularly timely study by Spratley, et al. 
presents this information in a creditable fashion. This study examines the 
"Location of Minority and Nonminority Health Professions," by geographic 
region and division; metropolitan vs. non-metropolitan area; county population 
size; situation in primary care manpower shortage areas; and poverty status of 
county residents. These analyses were accomplished for physicians, dentists 
and pharmacists. 

To better appreciate the distribution of minority health professionals. 
Table 28 was developed to show the relative availability of minority 
physicians to those minority populations previously described in selected 
counties. The patterns seen with these figures however suggest no clear 
trends across the States reviewed. 

Even through use of the Area Resource File, it has not been possible to obtain 
detailed information on the availability of these minority professionals to 
examine resources at the community level. Data on selected professions, 
notably medicine, will be most readily accessible and will be examined as its 
availability will permit. For example, preliminary reports on a study 
recently completed by the Rand Corporation, surveying 1,000 physicians who 
graduated in 1975 shows that the minority physicians were much more likely 
(11 percent versus 6 percent) to locate their practices in health manpower 
shortage areas, than nonminority physicians. 



456 



■feble 28 

Number of Physicians, Population and 

Riysician per 100,000 Population Ratios in 

Counties vd.th a Significant Minority Population 



1 
1 


Counties with 20% + Black Pop. [ 

[ 


1 1 
[ Counties vd.th 20% + Hispanic Pop. [ 

[ [ 


1 








Black [ 








Hispanic [ 


1 








Phys. [ 








Phys. 1 


1 








Per 1 








Per [ 


1 




No. of 




100,0001 




No. of 




100,000 1 


1 


No. of 


Black 


Riark 


Black [ 


[ No. of 


Hispanic 


Hispanic 


Hispanic [ 


1 state 


Counties 


Ftiys. 


Pop. 


Pop. [ 


[Counties 


Riys. 


Pop. 


Pop. 1 


1 

1 Alabama 


37 


219 


804,738 


27.2 1 










lArizona 


— 


— 


— 


— [ 


[ 8 


U9 


224,383 


53.0 1 


lArkansas 


27 


65 


318,388 


20.4 [ 


[ — 


— 


— 


— [ 


Irfllifomia 


— 


— 


— 


— [ 


1 11 


1,055 


2,697,924 


39.1 [ 


1 Colorado 


— 


— 


— 


— [ 


1 12 


20 


79,603 


25.1 [ 


iDist. of Colum. 


1 


467 


444,808 


105.0 [ 


[ — 


— 


— 


— [ 


1 Florida 


12 


64 


264,381 


24.2 [ 


1 1 


2,375 


580,025 


409.5 [ 


[Georgia 


106 


442 


1,303,220 


33.9 [ 


[ — 


— 


— 


— 1 


JHawaii 


— 


— 


— 


— [ 


[ — 


— 


— 


— [ 


1 Illinois 


4 


980 


1,414,734 


69.3 [ 


[ — 


— 


— 


— [ 


[Indiana 


2 


80 


279,680 


28.6 


[ — 


— 


— 


— [ 


[Kansas 


1 


14 


41,274 


33.9 [ 


[ — 


— 


— 


— [ 


[Kentucky 


1 





16,695 


0.0 1 


[ — 


— 


— 


— 1 


[Louisiana 


45 


195 


1,023,876 


19.0 [ 


[ — 


— 


— 


— [ 


[Maryland 


9 


443 


728,137 


60.8 [ 


[ — 


— 


— 


— [ 


[Michigan 


1 


479 


823,871 


58.1 [ 


[ — 


— 


— 


— [ 


[Mississippi 


64 


60 


766,449 


7.8 [ 


[ — 


— 


— 


— 1 


[MLssouri 


2 


98 


?n,263 


46.4 [ 


[ — 


— 


— 


— [ 


[Mmtana 


- 


— 


— 


— [ 


[ — 


— 


— 


— [ 


[ifebraska 


— 


— 


— 


— 1 


[ — 


— 


— 


— [ 


[New Jersey 


1 


210 


311,630 


67.4 [ 


1 1 


1?? 


145,249 


84.0 [ 


[NewMeadco 


— 


— 


— 


— 1 


1 28 


202 


449,749 


44.9 1 


[New York 


3 


937 


1,328,927 


70.5 [ 


1 2 


652 


730,385 


89.3 [ 


[North Carolina 


56 


292 


1,092,088 


26.7 1 


[ — 


— 


— 


— [ 


[North mkota 


— 


— 


— 


— [ 


[ — 


— 


— 


— [ 


[Ohio 


1 


203 


338,254 


60.0 1 


[ — 


— 


— 


— [ 


[Oklahuiu 


— 


— 


— 


— 1 


1 — 


— 


— 


— [ 


[Pennsylvania 


1 


181 


633,597 


28.6 1 


[ — 


— 


— 


— [ 


[South Carolina 


39 


128 


790,288 


16.2 [ 


[ — 


— 


— 


— [ 


[South EBkota 


— 


— 


— 


— [ 


[ — 


— 


— 


— [ 


[Teiuiessee 


8 


261 


496,114 


52.6 [ 


[ — 


— 


— 


— [ 


[Texas 


27 


76 


241,778 


31.4 [ 


[ 90 


1,238 


1,963,334 


63.1 1 


[Utah 


— 


— 


— 


— [ 


[ — 


— 


— 


— [ 


[Virginia 


49 


233 


748,580 


31.1 [ 


[ — 


— 


— 


— [ 


[Vfeshington 


— 


— 


— 


— [ 


[ 1 





2,950 


0.0 1 


[Wisconsin 
1 


— 


— 


— 


— ~ 1 


I ^^ 


' 


^^ 


" 1 


1 

[T«=! Angeles 


1 


1,1?3 


925,832* 


121.3 [ 


[ 1 


901 


2,065,503 


43.6 [ 


[San Francisco 
[ 


1 


U9 


84,334* 


141.1 [ 


[ 1 


110 


84,194* 


130.7 [ 



*The minority population group comprises less than 20% of the total population 
in these counties. 

457 



•feble 28 (continued) 

Number of Hiysicians, Population and 

Physician per 100,000 Population Eatios in 

Counties vd.th a Significant Minority Population 



1 
1 
1 


Counties with 20% + Indian Pop. | 

1 


1 1 
1 Counties with 5.0% + Asian Pop. | 

1 1 


1 








Indian | 








Asian | 


1 








Phys. 1 








Hiys. 1 


1 








Per 1 








Per 1 


1 




No. of 




100,0001 




No. of 




100,000 i 


1 


No. of 


Indian 


Indian 


Indian | 


1 No. of 


Asian 


Asian 


Asian | 


1 state 


Counties 


Phys. 


Pop. 


Pop. 1 


1 Counties 


Riys. 


Pop. 


Pop. 1 


1 

1 Alabama 


_ 


_ 














1 Arizona 


3 


8 


92,159 


8.7 1 


1 — 


— 


— 


— 1 


lArkansas 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


Inalifomia 


— 


— 


— 


— 1 


1 9 


3,025 


853,715 


354.5 1 


1 Colorado 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


IM-St. of Colum. 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Florida 


— 


— 


— 


— 1 




1 


— 


— 


— 1 


1 Georgia 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


iHawn't 


— 


— 


— 


— 1 


4 


642 


449,289 


142.9 1 


1 Illinois 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Indiana 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


iKansas 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


(Kentucky 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Louisiana 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Maryland 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Michigan 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Mississippi 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


iMissouri 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Montana 


5 





18,502 


0.0 1 


1 — 


— 


— 


— 1 


1 Nebraska 


1 





2,436 


0.0 1 


1 — 


— 


— 


— 1 


|New Jersey 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


[New Meadco 


3 


18 


73,452 


24.5 1 


1 — 


— 


— 


— 1 


iNew York 


— 


— 


— 


— 1 


1 2 


2,102 


168,061 


1750.7 1 


iNDrth Carolina 


2 


5 


38,000 


13.2 1 


1 — 


— 


— 


— 1 


1 North Dakota 


3 





11,638 


0.0 1 


1 — 


— 


— 


■ — 1 


johio 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


lOklahcnia 


3 


8 


19,157 


41.8 1 


1 — 


— 


— 


— 1 


1 Pennsylvania 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 South Carolina 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 South EBkota 


10 





28,885 


0.0 1 


1 — 


— 


— 


— 1 


1 Tennessee 
1 Texas 
iutah 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 





5,622 


0.0 1 


1 _^ 




..^ 


1 


[Virginia 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Washington 


— 


— 


— 


— 1 


1 — 


— 


— 


— 1 


1 Wisconsin 
1 


1 





3,014 


0.0 1 


1 — 


— 


— 


— 1 


1 

ILos Angeles 


1 


18 


52,809* 


34.1 1 


1 1 


2,095 


418,966 


500.0 1 


|San Francisco 
1 


1 





3,374* 


0.0 1 


1 1 


283 


144,619 


195.7 1 



*The minority population group comprises less than 20% of the total population 
in these counties. 

458 



Summary . The total minority representation in health occupations in 1980 was 
21.0 percent compared to a U.S. minority population of 20.1 percent. However, 
the minority percentages are heavily inflated by the large numbers of minority 
nursing aides, orderlies, and attendants (477,000), registered nurses 
(171,000), and licensed practical nurses (102,000). The percentages of 
minorities in professional health occupations, excepting nursing, are far 
below the minority percentages in the U.S. population. 

Likewise, in mental health and mental health-related occupations such as 
psychology, social work, and alcoholism and drug treatment, the professional 
staffs are overwhelmingly nonminority even when the clientele is substantially 
minority. 

The graphs on pages 82 to 85 show that there has been some recent improvement 
in the ratios to population of minority physicians, dentists, pharmacists, and 
R.N.s but this has only been in the context of a trend for all segments of the 
population. Relative to the nonminority population, the minority health 
professional ratios per 100,000 minority population have remained about the 
same. In Table 28, the gap between minority populations and the nonminority 
population is clearly illustrated. The total U.S. physician-population ratio 
is about 213 per 100,000. No physician population ratio for State aggregated 
counties having 20 percent or more Black or American Indian populations even 
approach 213. Most are below 100. There is one state (Florida) where the 20 
percent Hispanic counties' ratio exceeds the national ratio; all others are 
markedly lower. 

PART II. B. 2. - Development of Minority Health Professionals 

Because of the historic and documented continuing underrepresentation of 
minorities in the health professions, and the question of the current and 
future availability of professionals to serve minority communities, part of 
this study was focussed on the development of minority health professionals. 

As a step beyond looking at whether health professionals were available to 
minority communities, the subcommittee sought to briefly examine whether the 
education of health professionals seemed available to those same minority 
communities. If it has proven difficult to attract practitioners to many of 
these communities, perhaps success could be enhanced if the educational and 
training programs and facilities were both available and accessible to those 
minority communities. 

The purpose of this analysis was to document the degree to which the numbers 
of minority students in health professions training suggests progress in 
narrowing the gap of underrepresentation. It was also to examine the 
immediate availability of health professions training programs to the counties 
with significant minority populations, and the degree of success of those 
minority students in graduating from public versus private institutions. With 
the understanding that government (i.e. public schools) has a greater 
responsibility than the private sector for assuring equity and addressing the 
health and educational needs of all citizens, a review of public and private 
accomplishments was undertaken. 



459 



For the first phase of this review, an examination of U.S. trends in 
first-year minority enrollment for the various health disciplines is provided 
in graphic and tabular form. 

The most significant increase in Black first-year medical students occurred in 
the late 1960s and early 1970s when the percentage of Black first-year 
students rose from 2.7 percent to 7.5 percent. The percentage then fell to 
under 7 percent and remained at that level throughout the rest of the 1970s 
and into the 1980s. The increase in Black undergraduates similarly, peaked, 
fell, and then stabilized at a somewhat lower level. See Figure 1. 

Asian/Pacific Islander first-year enrollment of medical students was only 
slightly above their representation in the U.S. population until 1975-76. It 
then increased precipitously and is still increasing. It is between three and 
four times higher than the percentage of Asian/Pacific Islanders in the U.S. 
population, and much higher than the percentage of Asian/Pacific Islander 
undergraduates. See Figure 2. 

The Hispanic trend lines resemble those for Blacks although at lower levels. 
The differences are that Hispanic first-year enrollment continued to increase 
for a longer time before leveling off and that the U.S. Hispanic population is 
increasing more rapidly than the Black population. See Figure 3. 

The American Indian first-year medical school enrollment percentage exceeded 
the Indian percentage of the U.S. population in 1973-74. However within two 
years, the percentage declined to what would have been a more expected level 
based on previous first-year enrollments. The aberrant 1973-74 figure may 
have been due to a successful program in that single year to increase American 
Indian students. See Figure 4. 

For each of the last five years for which there are data, the Indian 
percentage of first-year medical school enrollment has been 0.4. See Table 29. 



460 



Figure 1 



Blacks as a Percent of the United States Population, of Undergraduate Students, and 
of First-Year Enrollees in Schools of Medicine: 1968-69 Through 1983-84 




• Percent of Population 

• Percent of Undergraduate Students 

• Percent of First-Year Enrollment 



1970-71 



'75-'76 
Academic Year 



'80-'81 



461 



Figure 2 



Hispanics as a Percent of the United States Population, of Undergraduate Students and 
of First- Year Enrollees in Schools of Medicine: 1968-69 Through 1983-84 



Percent 
7 , — 



»»♦♦ 



»*♦ 



»»♦♦ 



.»••* 



»»»' 



»»•» 



»♦*> 



»••• 



»•• 



.♦• 



o»**' 



• • 



,••' 



• • t 






— Percent of Population 

Percent of Undergraduate Students 

— Percent of First-Year Enrollment 



J I 



1970'71 



75-'76 
Academic Year 



'80-'81 



462 



Figure 3 

Asians as a Percent of the United States Population, of Undergraduate Students, and 
of First- Year Enrollees in Schools of Medicine: 1968-69 Through 1983-84 



Percent 
6 I — 




Percent of First-Year Enrollment 

Percent of Undergraduate Students 

Percent of Population 



1970-71 



75-76 
Academic Year 



'80-'81 



463 



Figure 4 



American Indians as a Percent of the United States Population, of Undergraduate Students, and 
of First- Year Enrollees in Schools of Medicine: 1968-69 Through 1983-84 



Percent 
0.8 



0.6 



0.4 



0.2 



.•' ••. 



'•■•••.....-••• : ,---'"' 

- ■; 




Percent of Undergraduate Students 

Percent of Population 

Percent of First-Year Enrollnnent 



J L 



1970'71 



'75-'76 
Academic Year 



'80-'81 



464 



Table 29 

First-Year Enrollment in Schools of Medicine in 

the United States, By Racial/Ethnic Category: 

Academic Years 1968-69 Through 1983-84 





TOTAL 
FIRST 








Racial/ethnic category 










NON- 






















MINORITY 




YEAR 


MINORITY 


UNDER- 
















FIRST 




ENROLL- 


FIRST 


REPRESENTED 


1 














YEAR 


Academic 


MENT 


YEAR 


MINORITIES 






Mainland 










ENROLL- 


year 




ENROLL- 






Mexican 


Puerto 


Other 


American 




Other 


MENT 




J./ 


MENT 


-2/ 


Black 


Anerlcan 


Rlcan 


Hispanic 


Indian 


Asian 


minority 














Number 


of students 












1968-69 


9,863 


413 


292 


266 


20 


3 


3/ 


3 


121 


3/ 


9.450 


1969-70 


10,422 


641 


501 


^440 


44 


10 


-T/ 


7 


140 


~J/ 


9.781 


1970-71 


11.348 


998 


808 


697 


73 


27 


"7/ 


11 


190 


~J/ 


10,350 


1971-72 


12,361 


1,280 


1.063 


882 


118 


40 


^/ 


23 


217 


-7/ 


11.081 


1972-73 


13,677 


1,437 


1.172 


957 


137 


44 


^/ 


34 


231 


3T 


12.240 


1973-74 


14,159 


1,631 


1.301 


1.027 


174 


56 


~J/ 


44 


259 


71 


12.528 


1974-75 


14,763 


1,839 


1.473 


1.106 


227 


69 


-7/ 


71 


275 


91 


12.924 


1975-76 


15,295 


1,787 


1,432 


1.036 


224 


71 


TT 


60 


282 


73 


13.508 


1976-77 


15,613 


1,891 


1,462 


1.040 


245 


72 


62 


43 


348 


81 


13.722 


1977-78 


16,136 


2,002 


1,607 


1.085 


246 


68 


157 


51 


395 


3/ 


14.134 


1978-79 


16,501 


2.046 


1,594 


1.061 


260 


75 


151 


47 


452 


~J/ 


14.455 


1979-80 


16,930 


2.237 


1,735 


1.108 


290 


86 


188 


63 


502 


"1/ 


14.693 


1980-81 


17,186 


2,344 


1.772 


1.128 


258 


95 


224 


67 


572 


"7/ 


14,842 


1981-82 


17,268 ,, 


2,683 


1.918 


1.196 


300 


105 


247 


70 


765 


~J/ 


14.585 


1982-83 


17,245 */ 


2,840 


1.904 


1.145 


305 


114 


278 


62 


936 


-7/ 


14.405 


1983-84 


17,146 H/ 


2.889 


1.906 


1.173 


301 


109 


248 


75 


983 


3/ 


14.257 












Percent 












1966-69 


100.0 


4.2 


3.0 


2.7 


0.2 


* 


3/ 


* 


1.2 


3/ 


95.8 


1969-70 


100.0 


6.2 


4.8 


4.2 


0.4 


0.1 


1/ 


0.1 


1.3 


3/ 


93.8 


1970-71 


100.0 


8.8 


7.1 


6.1 


0.6 


0.2 


"7/ 


0.1 


1.7 


"?/ 


91.2 


1971-72 


100.0 


10.4 


B.6 


7.1 


1.0 


0.3 


-1/ 


0.2 


1.8 


■7/ 


89.6 


1972-73 


100.0 


10.5 


8.6 


7.0 


1.0 


0.3 


"7/ 


0.2 


1.7 


03 


89.5 


1973-74 


100.0 


11.5 


9.2 


7.3 


1.2 


0.4 


~J/ 


0.3 


1.8 


0.5 


88.5 


1974-75 


100.0 


12.5 


10.0 


7.5 


1.5 


0.5 


■7/ 


0.5 


1.9 


0.6 


B7.5 


1975-76 


100.0 


11.7 


9.4 


6.8 


1.5 


0.5 


03 


0.4 


1.8 


0.5 


88.3 


1976-77 


100.0 


12.1 


9.4 


6.7 


1.6 


0.5 


0.4 


0.3 


2.2 


0.5 


87.9 


1977-78 


100.0 


12.4 


10.0 


6.7 


1.5 


0.4 


1.0 


0.3 


2.4 


3/ 


87.6 


1978-79 


100.0 


12.4 


9.7 


6.4 


1.6 


0.5 


0.9 


0.3 


2.7 


"7/ 


87.6 


1979-80 


100.0 


13.2 


10.2 


6.5 


1.7 


0.5 


1.1 


0.4 


3.0 


3/ 


86.8 


1980-81 


100.0 


13.6 


10.3 


6.6 


1.5 


0.6 


1.3 


0.4 


3.3 


3/ 


86.4 


1981-82 


100.0 


15.5 


11.1 


6.9 


1.7 


0.6 


1.4 


0.4 


4.4 


3/ 


84.5 


1982-83 


100.0 */ 


16.5 


11.0 


6.6 


1.8 


0.7 


1.6 


0.4 


5.4 


3/ 


83.5 


1983-84 


100.0 3/ 


16.8 


11.1 


6.8 


1.8 


0.6 


1.4 


0.4 


5.7 


3/ 


83.2 



* Less than 0.05 percent. 

1/ Residents of the Comnonwealth of Puerto Rico are not considered to be members of any minority group and are 

~ Included In this table only In the TOTAL FIRST-YEAR ENROLLMENT and the Mainland Puerto Rlcan data columns. 

_2/ Includes all minority racial/ethnic categories except Asian and Other minority. 

_3/ The categories "Other Hispanic" and "Other alnorlty" were not In use in these years. 

4/ Excludes 9 students for «haa racial ethnic Inforaatlon was not available. 

5/ Excludes 4 students for wtioa racial ethnic Information was not available. 



465 



Prior to 1972-73, the number of minority first-year LPN enrollment was greater 
than the number of minority first-year RN enrollment. Since then, the RN 
enrollment number have continued to rise while the LPN enrollment number fell 
for several years and then began to increase slightly at the end of the 
1970s. See Figure 5. 

The percentage of minority first-year LPN enrollment has always remained above 
that for RN students. However, in the last several years the gap has been 
closing. See Table 30. 



466 



Figure 5 



0) 

ni 

"8 



-) 






O 


03 


00 


f 


(T) 


+^ 




C 


^ 




m 


E 


3 
O 


m 


k. 


w 


s: 


a: 


+-' 


O 


CN 


Q. I>. 



Oi 



•n 

c 



ro 


03 


z> 


cc 


7, 


c: 


4^ 




u 




0} 


c 


1) 


V 


V) 


h 




o 


in 

C 


c 


o 


LU 


t/s 




w 


^ 


0} 


fn 


« 


09 

> 


o 


■t-< 






Tl 




C 


u. 


CO 


> 




+- 




^ 


CD 



:c/3 



i-8 

«3 



0) S 

•n c 

c ■- 
ni 

0) 

E 

3 
Z 





• 


\ 






"~ 




• 


X 










• 


\ 






_ 




• 


X 










• 


X 










• 


X 








t/» 


• 


X 






— 


E 


• 
• 


\ 




E 




11 


• 
• 


\ 




ra 








m 




O 




cii 


• 


# 


CI 






IE 


• 


M 


o 


o. 




• 


i 


c 


Z 


~ 


2-1 


• 


# 


5a: 




1 


• 


M 








1 


• 


M. 


1 




— 




X 










• 


# 










• 


# 






_ 




• 


% 










• 


X 








• 


X 








• 
• 




X 








• 




X 








• 




X 








• 




X 








• 




X 






^ 


• 




X 








• 




X 








• 




% 








1 — * I 




1 


i 







o 

CN] 




467 



Table 30 



FIRST-YEAR ENROLLMENT IN REGISTERED NURSE AND PRACTICAL NURSE TRAINING PROGRAMS IN THE UNITED STATES AND POSSESSIONS. 
BY TYPE OF PROGRAM AND RACIAL/ETHNIC CATEGORY: SELECTED ACADEMIC YEARS 1962-63 THROUGH 1980-81 



















V 








TOTAL 










Racial /ethnic 


category 








First-year 


















FIRST-YEAR 


n 


Inorlty 
















Academic 


ENROLLMENT 


enrollment 


Black 




Hispanic 


Other 


minority 


year 
























Percent 




Percent 






Percent 




Percent 








of total 




of total 






of total 




of total 








first-year 




first-year 






first-year 




first-year 






Number 


enrollment 


Number 


enrollment 




Number 


enrollment 


Number 


enrollment 










All 


registered nurse 


programs 


» 






1962-63 


48,259 


* 


* 


1,456 


3.0 




* 


* 


* 


* 


1965-66 


59,049 


* 


* 


1,891 


3.2 




* 


* 


* 


* 


1968-69 


60.598 


• 


* 


3,735 


6.2 




* 


* 


* 


* 


1971-72 


91,896 


9,889 


10.8 


7,088 


7.7 




1,866 


2.0 


935 


1.0 


1974-75 


89,706 


11,322 


12.6 


8,159 


9.1 




2,080 


2.3 


1.083 


1.2 


1977-78 


101,438 


11,212 


11.2 


7,313 


7.2 




2.520 


2.5 


1,379 


1.4 


1980-81 


102,540 


13.799 


13.5 


8.537 


8.3 




3,515 


3.4 


1,747 


1.7 










RN baccalaureate 


degree programs 








1962-63 


8,867 


* 


* 


433 


4.9 




* 


* 


* 


* 


1965-66 


11,590 


* 


* 


554 


4.8 




* 


* 


* 


* 


1968-69 


14,111 


* 


* 


842 


6.0 




* 


* 


* 


* 


1971-72 


26,758 


3,509 


13.1 


2.407 


9.0 




667 


2.5 


435 


1.6 


1974-75 


29,479 


4,911 


16.7 


3.650 


12.4 




807 


2.7 


454 


1.5 


1977-78 


35,442 


4,366 


12.3 


2.905 


8.2 




970 


2.7 


491 


1.4 


1980-81 


32,548 


5,390 


16.6 


2.797 


8.6 




1,813 


5.6 


780 


2.4 










RN 


associate degree 


programs 








1962-63 


3,317 


* 


* 


173 


5.2 




* 


* 


* 


* 


1965-56 


8,555 


* 


* 


558 


6.5 




* 


* 


* 


* 


1968-69 


17,808 


* 


* 


1.871 


10.5 




* 


* 


* 


* 


1971-72 


35,863 


4,958 


13.8 


3,550 


9.9 




1.034 


2.9 


374 


1.0 


1974-75 


38,581 


5,096 


13.2 


3,495 


9.1 




1,069 


2.8 


532 


1.4 


1977-78 


46,755 


5,515 


11.8 


3,580 


7.6 




1,318 


2.8 


617 


1.3 


1980-81 


53.127 


6,993 


13.2 


4,668 


8.8 
RN diploma 


1 pro* 


1,477 
grams 


2.8 


848 


1.6 


1962-63 


36,075 


* 


* 


850 


2.4 




* 


* 


* 


* 


1965-66 


38,904 


* 


* 


779 


2.0 




* 


* 


* 


* 


1968-69 


28,679 


* 


•* 


1,022 


3.6 




* 


* 


* 


* 


1971-72 


29,275 


1.422 


4.9 


1,131 


3.9 




165 


0.6 


126 


0.4 


1974-75 


21 ,646 


1.315 


6.1 


1.014 


4.7 




204 


0.9 


97 


0.4 


1977-78 


19,241 


1.331 


6.9 


828 


4.3 




232 


1.2 


271 


1.4 


1980-81 


16,865 


1,416 


8.4 


1.072 


6.4 




225 


1.3 


119 


0.7 










All 


practical nurse 


programs 








1962-63 


27,085 


* 


* 


4.455 


16.4 




* 


* 


* 


* 


1965-66 


36,768 


* 


* 


6,669 


18.1 




* 


'* 


* 


* 


1968-69 


44,917 


* 


* 


7,804 


17.4 




* 


«■ 


* 


* 


1971-72 


57,567 


11,183 


19.4 


8,545 


14.8 




1,965 


3.4 


673 


1.2 


1974-75 


45,530 


8,313 


17.9 


5,795 


12.5 




1,927 


4.1 


591 


1.3 


1977-78 


53,002 


8,279 


15.6 


5,883 


11.1 




1,655 


3.1 


741 


1.4 


1980-81 


51 ,335 


8,993 


17.5 


6,252 


12.2 




2,010 


3.9 


731 


1.4 



* Data for minorities other than black were not collected until 1971-72. 

_y Data for academic years 1962-63 through 1968-69 are based on those first-year students In schools responding to 
question on minority enrollment; data for 1971-72 through 1980-81 are based on those students In schools 
responding to question on minority enrollment, aale enrollnent. or both. 



468 



In the second phase of the review, a comparison was made of the minority 
graduates (by racial/ethnic group) from health professions schools in counties 
having greater than 20 percent of that minority versus schools elsewhere in 
the State. Medical, dental and pharmacy schools were analyzed. A series of 
tables have been prepared displaying this information and are provided in 
Appendix II. B. 1. A discussion of what those tables display is included in 
this part. In addition, there are four tables comparing percentages of public 
and private medical, dental, and pharmacy school graduates for the four 
racial/ethnic groups considered in this report provided in Appendix II. B. 1. 

A comparison of minority graduates from health professions schools in counties 
having greater than 20 percent of that minority versus schools elsewhere in 
the same States. 

Black graduates . Although 51 U.S. medical schools are in counties with over 
20 percent Black populations, in no county except where Howard University and 
Meharry Medical College are located, do Black medical school graduates exceed 
14 percent. The average percentage of Black graduates from the 51 schools is 
7.4. When Howard and Meharry are excluded, the average is 5.0 percent. From 
the 35 medical schools not in counties with 20 percent and over Black 
populations, the percentage of Black graduates is 3.8. The number of Black 
graduates in the 20 percent and over Black population counties is 510 (76.8 
percent) compared to 154 (23.2 percent) in the other counties. Of the 51 
medical schools, 18 (35.3 percent) are public. Of the 35 schools, 30 (85.7 
percent are public. Private schools are more likely to be located in areas 
with Black communities and to have, overall, higher percentages of Black 
graduates than public medical schools in the States being considered. See 
Table 31. 



Table 31 
Black Medical Graduates 



1 Location of 


Number of 
Schools 


Total 
Graduates 

6,898 

4,071 


Number of 

Minority 

Graduates 

923 

549 


Number of | 
Black 1 


1 School 


Public Private 
18 33 
30 5 


Graduates | 


1 Counties 20%+ Black 
1 Population 

1 Counties under 20% 
1 Black Population 


510 1 
154 1 



469 



A similar pattern exists for dental schools . Both Howard and Meharry have 
greater than 50 percent Black graduates, but in only one other county (that of 
the Medical College of Georgia) are there more than seven percent Black dental 
graduates. The average percentage of Black graduates from the 26 schools in 
counties with 20 percent and over Black population is 5.4. If Howard and 
Meharry are not counted, the percentage is 2.5. It is 2.0 percent in the 14 
dental schools in all other counties in the same States. In the 20 percent 
and over Black population counties, there are 145 (82.9 percent) Black 
graduates but only 30 (17.1 percent) in all other counties. This difference 
is largely a reflection of the graduate totals for Howard and Meharry. Twelve 
(46.1 percent) of 26 schools in the 20 percent and over counties and 11 (78.6 
percent) of the 14 other schools are public. When Howard and Meharry are 
eliminated, the percentage differences between public and private dental 
school graduates are also eliminated. See Table 32. 



Table 32 
Black Dental Graduates 





Number of 




Number of 


Number of | 


1 Location of 


Schools 


Total 
Graduates 


Minority 
Graduates 


Black 1 


1 School 


Public Private 


Graduates | 


1 Counties 20%+ Black 




1 Population 


12 14 


2,678 


324 


145 1 


1 Counties under 20% 










1 Black Population 


11 3 


1,559 


147 


30 1 



470 



J 



For 22 pharmacy schools in the 20 percent and over Black counties, the average 
percentage of Black graduates is 6.7. In all other counties the average of 17 
schools is 2.9 percent. Three predominantly Black schools (Florida A & M 
University, Howard University and Xavier University) are in the 20 percent and 
over counties, and a forth predominantly Black school (Texas Southern 
University) is in an under 20 percent Black county. The number of Black 
pharmacy graduates in the 20 percent and over counties is 146 (71.9 percent) 
and 57 (28.1 percent) in the other counties in the same States. Of the 22 
schools in over 20 percent counties, 13 (59.1 percent) are public. In all 
other counties, 13 of 17 (76.5 percent) are public. Differences in 
percentages of graduates between public and private pharmacy schools are 
attributable to relatively large number of graduates from the public 
predominantly Black institutions. See Table 33. 



Table 33 
Black Pharmacy Graduates 



1 Location of 


Number of 
Schools 


Total 
Graduates 

2,189 

1,956 


Number of 

Minority 

Graduates 

350 

201 


Number of I 
Black 1 


1 School 


Public Private 
13 9 
13 4 


Graduates | 


1 Counties 20%+ Black 
1 Population 

1 Counties under 20% 
1 Black Population 


146 1 

57 1 



471 



Hispanic graduates . In counties with 20 percent and over Hispanic 
populations, there are six medical schools graduating 7.1 percent Hispanics. 
The figure is 7.5 percent for 16 schools outside those counties. The reason 
is that several schools in California not in Hispanic counties but close to 
them enroll fairly sizable numbers of Hispanic students. The number of 
Hispanic medical students graduating from schools not in areas defined by this 
report as Hispanic communities, is 143 (71.1 percent) versus 58 (28.9 percent) 
in 20 percent and over Hispanic counties. The medical school catchment area 
for Hispanic students extends beyond their immediate communities, particularly 
in California. The proportions of public and private schools in both the over 
and under 20 percent Hispanic counties are about the same - 65-75 percent 
public. The public medical schools in California and New Mexico have the 
highest percentages of Hispanic graduates. The percentages of private medical 
schools, except for the University of Miami, are much lower. See Table 34. 



Table 34 
Hispanic Medical Graduates 





Number of 




Number of 


Number of | 


1 Location of 


Schools 


Total 
Graduates 


Minority 
Graduates 


Hispanic | 


1 School 


Public Private 


Graduates | 


1 Counties 20%+ 




1 Hispanic Population 


4 2 


816 


134 


58 1 


1 Counties under 20% 










1 Hispanic Population 


13 3 


1,911 


298 


143 1 



472 



There are only three dental schools in 20 percent and over Hispanic counties. 
Their average percentage of Hispanic graduates is 10.3 which is four points 
higher than for the eight dental schools in other counties in the same 
States. There were only 88 Hispanic dental graduates in the year being 
analyzed. Fifty (56.8 percent) were from the eight schools not in Hispanic 
communities and 38 (43.2 percent) were from the three schools in 20 percent 
and over Hispanic counties. Four public dental schools had percentages of 
Hispanic graduates between ten and sixteen percent. No private dental school 
had more than 4.5 percent Hispanic graduates. See Table 35. 



Table 35 
Hispanic Dental Graduates 





Number of 




Number of 


Number of | 


1 Location of 


Schools 


Total 
Graduates 


Minority 
Graduates 


Hispanic | 


1 School 


Public Private 


Graduates | 


1 Counties 20%+ 




1 Hispanic Population 


2 1 


368 


74 


38 1 


1 Counties under 20% 










1 Hispanic Population 


5 3 


774 


117 


50 1 



Four Hispanic graduates of pharmacy schools , the pattern is almost identical 
to that for dental graduates. Hispanics are 9.1 percent of graduates in 20 
percent and over Hispanic counties and 7.5 percent in all other counties. 
There are four schools in the Hispanic counties and eight outside them. The 
number of graduates in Hispanic counties is 26 (31.0 percent) and 58 (69.0 
percent) in other counties. Public pharmacy schools in Texas and New Mexico 
have the highest percentages of graduates and private schools are lower. See 
Table 36. 

Table 36 
Hispanic Pharmacy Graduates 



1 Location of 


Number of 
Schools 


Total 
Graduates 

287 

769 


Number of 

Minority 

Graduates 

107 

198 


Number of | 
Hispanic | 


1 School 


Public Private 
3 1 
7 1 


Graduates | 


1 Counties 20%+ 

1 Hispanic Population 

1 Counties under 20% 
1 Hispanic Population 


26 1 
58 1 



473 



American Indian graduates . There are so few American Indian graduates of 
medical, dental and pharmacy schools that the comparison by type of county has 
no meaning. There are no medical, dental or pharmacy schools in counties with 
20 percent or more American Indian population. See Table 37. 



Table 37 







Number of | 




Number of Number of 


American | 


1 Location of 


Schools Total Minority 


Indian | 


1 School 


Public Private Graduates Graduates 
American Indian Medical Graduates 


Graduates | 






1 Counties 20%+ 






1 American Indian 






1 Population 





1 


1 Counties under 20% 






1 American Indian 






1 Population 


9 4 1,296 126 
American Indian Dental Graduates 


15 1 


1 Counties 20%+ 






1 American Indian 






1 Population 





1 


1 Counties under 20% 






1 American Indian 






1 Population 


2 2 296 9 
American Indian Pharmacy Graduates 


1 


1 Counties 20%+ 






1 American Indian 






1 Population 





1 


1 Counties under 20% 






1 American Indian 






1 Population 


2 2 699 41 


5 1 



Note: For Blacks, Hispanics, and American Indians, "minority communities" 
are defined as counties with 20 percent or greater minority 
population. For Asian/Pacific Islanders, the definition is counties 
with 2.0 percent Asian/Pacific Islander population and over. 



474 



J 



Asian/Pacific Islander graduates . The comparison here uses 2 percent of the 
population as a benchmark because there are so few U.S. counties with sizable 
numbers of this minority group. It also only involves counties in the States 
of Hawaii, New York and California. There are more Asian/Pacific Islander 
graduates from the schools in counties with 2 percent and higher Asian/Pacific 
Islander populations. However, for all three disciplines, the location of 
only one or two schools with high Asian/Pacific Islander enrollment in the 
counties accounts for the difference. See Table 38. 

Table 38 









Number of 








Asian/ 




Number of 


Number of 


Pacific 


1 Location of 


Schools 


Total Minority 
raduates Graduates 

Medical Graduates 


Islander 


1 School 


Public Private G 
ian/Pacific Islander 


Graduates 


1 As 




1 Counties 2.0%+ 








1 Asian Population 


3 6 


1,327 250 


127 


1 Counties under 2.0% 








1 Asian Population 


9 1 


1,276 164 


59 


1 As: 


Lan/Pacific Islander 


Dental Graduates 




1 Counties 2.0%+ 








1 Asian Population 


2 3 


611 84 


28 


1 Counties under 2.0% 








1 Asian Population 


2 2 


346 51 


34 


1 Asian/Pacific Islander 


Pharmacy Schools 




1 Counties 2.0%+ Asian 








1 Population 


1 2 


381 138 


104 


1 Counties under 2.0% 








1 Asian Population 


3 1 


584 80 


54 



Summary ; Overall, these comparisons shows that minority health professions 
graduates are more likely to be graduated from schools located in minority 
communities (counties), but even in these counties the percentages of minority 
graduates are still much lower than the minority representation in the total 
county population. Asian/Pacific Islander graduates do not conform to this 
pattern because they are so overrepresented among health professions graduates 
relative to their proportion of the total population and because there are so 
few large communities of Asian/Pacific Islanders. Subpopulations of 
Asian/Pacific Islanders are differently represented in the health professions 
and the total overrepresentation may disguise some underrepresentation of 
Asian/Pacific Islander subgroups within specific health professions. 



475 



For both Blacks and Asian/Pacific Islanders, there are higher percentages of 
graduates from private medical, dental, and pharmacy schools than from public 
schools. 

The only public health professions schools graduate percentages which exceed 
those for private schools are Hispanic dental graduates and American Indian 
pharmacy graduates. Among individual States, there are examples of higher 
percentages of public school graduates but this would be expected given the 
small number of schools in most States. 

There are private, predominantly Black health professions schools which 
influence the Black public-private ratio but there are no private, 
predominantly Hispanic, Asian/Pacific Islander or American Indian health 
professions schools. The data indicate that public health professions schools 
in the U.S. are generally not recruiting and graduating minority students as 
well as private schools. 

PART II. B. 3. - Practice Patterns of Minority Health Professionals 

Information on the practice patterns of minority health professionals was 
sought as an aid in defining the specific role and contribution of these 
individuals in delivering health care to minority communities. Although 
information was sought on a variety of health disciplines, the only detailed 
information obtained addressed the field of medicine, particularly Black 
physicians. Efforts will continue to identify such information with respect 
to other professions and other minority groups. Nonetheless, the analysis of 
the available data is presented below. 

There was only limited information available for review relevant to physician 
location intentions . Most notable was a study conducted by the Bureau of 
Health Manpower a decade ago, looking at the "Influence of Preceptorship and 
Other Factors on the Education and Career Choices of Physicians." In 
examining the preferences for practice location of medical students and 
residents who were engaged in preceptorship programs, the findings included 
the following: 

• The probability of preferring an inner-city practice location was 
highest for those who attended high school in an inner-city community, 
had average or less than average financial support from family or 
savings,... (and) were minority. 

• The probability of preferring a rural or small town practice location 
was highest for residents who attended high school in a rural or small 
town community, had less than average financial support from family or 
savings,... (and) were nonminority. 

• The probability of preferring another urban/suburban location was 
highest for residents who attended high school in an urban or suburban 
community, had above average financial support from family or 
savings,... (and) were nonminority. 



476 



The predisposition of these future doctors to prefer a probable practice 
location similar to that of the community in which they were raised was not 
surprising; nor was the probable influence of their racial/ethnic and family 
socioeconomic background. This does suggest evidence to support the 
potentially positive outcomes of training more minority health professionals, 
and persons from rural areas and small towns, to help meet the needs of 
underserved inner city and rural areas. 

Upon completion of medical school, the initial indicator of the future 
specialty practice plans of a physician is the postgraduate or residency 
training that is sought. The length of this additional training varies with 
the nature of the specialty selected, but generally requires three to five or 
more years. Table 39 shows the Racial/Ethnic Background of Residents on Duty 
as of September 1, 1981, 1982 and 1983. 



Table 39 

Ethnic Background of Residents on Duty 

September 1, 1981, 1982, and 1983 



1 Ethnic Background 


1981 


1982 


1983 1 


1 Black (non-Hispanic) 


3,472 


3,307 


3,379 1 


1 American Indian or 








1 Alaskan Native 


197 


152 


111 1 


1 Mexican-American 


714 


697 


743 1 


1 Puerto Rican 


1,223 


1,227 


1,343 1 


1 Other Hispanic 


1,414 


1,396 


1,587 1 


1 Asian/Pacific Islander 


6,468 


5,762 


5,632 1 


1 White (non-Hispanic) 


53,196 


55,417 


58,576 1 


1 Total 


66,684 


67,958 


71,371 1 



In a more detailed analysis (Graettinger and Swanson) covering 1977 through 
1983, the National Resident Matching Program and the Association of American 
Medical Colleges collaborated to provide the data on numbers and percentage 
distribution of the residents, by specialty, by gender and by racial/ethnic 
group, as shown in Tables 40 and 41. Separate figures are not available for 
Asian/Pacific Islanders or American Indians, the former because they were not 
deemed underrepresented and the latter because their scarce numbers (fewer 
than 50 per year) would not lend to meaningful separate analysis. 



477 



u-1 n 



v£) CM 












^00 



00 Q 

3^ 



in ,-{ 






00 Q 









ro 0^ cyi i^ 



vD en 00 ^r ^ 

O rH 



ro <f r^ r^ Q CN ro 

"" ■ VO en rH 



lO m rH 






vO O rH ro en iH iH 

Lo en csj 



^csocN m^om ino^rH 

0\enrH M-vTrH CNliH 

vo m ^r 



ON C» CN O 

lO CN 

C-1 



0^ ^J^ 0^ I^ 
csl 00 iH 
1^ 



en 00 en CN 



r^ lO <!■ rH 
VO l~~ CN t-l 
00 



Sj 



a 



1^ 

I 

to 






CO 

9 



00 ~d- 

\0 CN 

m 



in .rf 
CM -3- 



in -* 

>d- CN 

en 






rH CTN 

-d- 



CN 



J^ 






en in 
in 



o en 



en r^ 



Q "-I 



iH m 

CT\ '^l 

en rH 
m 



CN m 

ON o\ 

ON 



00 Q 



CN o 
en rH 



r-~ 00 



r^ CN 

en CN 



m -a- rH CN 
vD r^ 3 
in 



s 



m -* 



CN 



CSI 



00 en rH 



r^ini^ enmenoo cM^fvo 

CN rH ON -* r-l m rH 

•<]• en 



csi^r~-r-~ ON-^tmm oovOrHrH 
CN vo r^ -T 

m CN 



■* r-f 



en 



en CN 
-* CT^ 

ON 



cNvooN^ r-^ CO o^ irn cNvocncN 
rH en Q in ,-{ ^ 

r~ iTn en 



^r^r^vo -d-m<r<f enmrHrH 
CN vo r^ -* r--. rH 

m CM rH 



CM CN m -^r 

in VD CN rH 



C3N CN CN en 

00 r-{ 



o en r^ CM 

in I^ rH rH 
vO 



ON rH CN 



m ON 



en 1^ 

OS m 

<!■ rH 

en 



m 00 

00 



as 



CN m 
en 



in 
en 






CTN ^ 



r^ en 

CN r-\ 



rHenvoen m^g-oorH 

en -vT rH 00 ~3- 



B\^ 



CN 

m 



^ -a- 



00 CN 
ON >. 
m 



en >£> 

00 r-{ 
Csl 



^a 



C3N Q r^ en 
ON ^E 

■J- CN 



SR 



m en 



o en 



^9 



vO 



^ 



en 



s 



rH r-; 00 en CO 

00 m i?N 

<F en 



CMincM VO-*CMO ON 

m -^ tn o 



en en rH <• vS rH 



Si 


>£) 


CN 


r^ 

S 


^ 


^ 


t^ 


s 


CN 


O 


CN 


s 


m 


<N 


^ 


s 


o 


1 


?2 


S 


r~ 


s 


CM 


CN 


S5 


S 


r-i 


O 


m 

rH 


r-i 


Cvl 


1 


ON 


s 


CM 



ti 



d 




d 




^ 



CO 



s 

W 






478 



Oi-tOO OrOcsJO a^-^■(?-|C^J u-irOOi-l CNlvO-cfO 

<M CN lOt-H u-i O est 

■H 1-1 CNl 



r-i a\ <t r-i rooino cNjiTioofM CNicsioo r^<-osoo 

~' ' '^ rn 



en 



CM rH 



a 



g\"^Ot-l CvJ^OiHi-l OQONCNCM r-~000 O^CMO 

CN Q -T iH -^ in C^ 



\Or^fOrH rocsl-^iH -<d-^000 -*rO>HO r-HCN-o-rO 



iH rH M 



CNl ~a- ,H 

OS 



^ 



M 



1 



^^ 


§ g 


1^ 


^ti 


4-1 


^ 


Sp 


•^ 



CA 






(NOrHO QCNJ,HO Or^.HO 1-I.Hr-lrH CNir~.rOt-l 

CM 1-^ -^ ^ ro CNl 

<-\ r-t CNl 

qjCTi.— ICM C^OOO CNicXJr-^ro or^Oi— I CNirHOOCM 

LO p r;; <^ CJ^ rH <f 

CO CS) LTl CTs 



nrOOO cocOcOiH >OCOCNii-l iHCNJMO ooo^mo 

CNl r~» rn co r^ CN 



,— :<r-<j-ro ON-^cMO rHt^r^co --a-r^oo ocniiO'H 
~a- .H -^"H OS a^cot-i 

en CNl in 00 



CMOOO cn^CMO 0>0>c00 <rcNiOO <|-vOOCNl 

cN a\ o CNl <yi iH 



<f O vO i-l 


r^ CNl •<)• CNl 


r^ vD ^a- CO 


CN r-{ 


CNl 


00 <-i 


CO 


CM 


<r 



VO O iH 



<r CO CO 

rH <!■ i-l 
00 



CNirOOO <rCN100 iHi-HCNO .Hr-lOO lOONr-IO 

tH r- 0^ 1-1 CO -- . 



r^ 00 CNl iH 
00 

CNl 



UO vO CO O 

iH 

CN 






00 iH 



CM CO O O CN CTi ■<!■ 

3 Sg'^^ 



iHCOOO m<3-00 iHOOOO CNiHOO vDr~>d- 

rH m >^ CN £^ 

Sa^CNO uOCOrHO CMO"0t-( <r~d-CNO CM-*v£) 

O CN 1— I o^ vQ <r 1— I 

iH CN CO vB 



d 






fi 



479 









iH 




8 


S8^ 


Eo 


8 


^ 


:J 


^ 


8 


^ 


CO 


CN 

CO 


8 


m 


s^ 


in 
m 


8 




s 


f^8 






S d'3 


?^ 


<f 


?i 


R 


S 


S 


tH 


a 


a 


?5 


ON 


CN 


da 


in 


CO 


r~ 


<r 


r^ 


r^ 


CO o 




rW 


2 o- 


^ 


3 


R 


ON 


S 


■vt- 
iH 




a 


3 


CO 
CO 


ON 
CN 


:? 




5 


a 


^ 


CO 
ON 


a 


g 


CN 


s 


E^?5 






^ a 


ON 




3 


CO 
CO 


^ 


CO 


00 
CN 


vD 


r^ 


00 


3 


-<r 


v£> 


r~. 


^ 


CO 


CN 


-* 


<r 


NO 


s 


dS 






Z CO 


^ 


s 


« 


sa 


LO 


s 


^ 


a8§8 


S 


CO 

NO 


PJ 


m 

CN 


?^ 


3 


Ri 


^ 


?i 


3 


NO m 

CO CM 






1 ad 


3 


vO 


p^ 


CO 


^ 


Pi 


rH 


a 


m 

CN 


CN 


ON 


ON 


3 


NO 


NO 


in 


-H 


ON 


NO 


in 


LO NO 




.H 


§ c^ 


S3 


S 


s 


r^ 

^ 


R:S 


5^ 


f?i 


s^ 


CM 

ON 


CN 




3 


CN 
ON 


3 


Eo 


a 


m 
in 


a 


ON 
ON 


^ 


R§ 






'^ a 


ON 


CM 


a 


CO 
CO 


^ 


CSI 


^ 


^3 


o 


m 


<!• 


-* 


a 


in 


3 


CO 


CM 


CO 


CnI 


NO 


a 


;2ia 






H CO 




5 


8 




sj; 


CO 
CO 


8 


3 


^ 


in 


8 


m 

CM 


s 


S 


8 


a 


5! 


a8 


?^ 


^ 


a8 




^ 


§ d 


r>« 


?5 


m 


^ 


52 

CO 


rH 


S 




?i 


CM 
CM 


a 


ON 


a 


o 

rH 


S 


NO 


CO 


NO 


o 


r^ 


in 


NO m 


1 


iH 


Z CN 


ON 


O 




a 


CO 


ON 


a 


SdS 


?2 


s 


a 


s 


m 

m 


:S 




C3N 


CM 

00 


d 


8 


cO ~* 




" P 


ON 


^a 


^ 


CO 


o 

CN 


a 


vO 


v£) 


ON 


a 


<r 


a 


00 


3 


CO 


CO 


CO 


rH 


r^ 

^ 


;5 


a« 






Z CO 


ON 


^ 


;2; 


s 




ON 


tn 


S 


00 


in 

ON 


8 


in 


!n 


ss 


S 


SRIS 


[^ 


5 


ON 


af^ 


3 




r^ 

r^ 
■^ 


T-l 




^ 
s 


in 


rH 
CM 


s 


ON 


rH 
Csl 




m 

CM 


ON 

§8 


5 
^ 


a 


rH 
NO 


3 


in 

CNj 


CN 

ON 


CO 


NO CO in o 

NO o o m 

O nD On O 






" ;5 


o 


^ 


00 


^^ 


CO 
CN 


?8 


^ 


r^ 


m 


3 


<f 


a 


m 


m 


CO 


CO 


r-^ 


CO 


NO 


a 


3?1 






1 S^ 


5 




3 


CO 


^ 




a 


s 


^ 


^ 




S 


s 




a 


S3 

On 


?d3!; 


m 
m 


3 


a§^8 


^ 
9 


1 d 


s 


?^ 


a 


CO 


^ 


rH 


a 


a 


?3 


R 




ON 


in 

iH 


d 


a 


in 


in 


CM 


<t 


r«. 


CO 


cN cy\ 


6 


z m 


?e 


?5 




2 


rH 
ON 


8 


?! 


UO 
Csl 


^ 


?^ 


;2i 


55 


?^ 


NO 


?R 


CO 


8 


So 


NO 


S8 


ON 


oQ r^ 

-* NO 




T— 1 


d 


CM 


3 


^ 


CN 

CO 


R 


?^ 


r~ 


ON 


a 


r~ 


o- 


ON 


m 


CN 


CO 


m 


CO 


<!• 




in 

rH 


aa 


1 




1 ^ 


S 


CNl 


gj 


2? 
o\ 


CM 

iH 


CN 

CO 




s 


rH 


CM 

ON 


CO 


SSd 


CO 


00 
CN 


S 


5* 


8 


NO 


a 


r1?5 


00 

ON 


1 dd 


r-l 
CM 


3 


rH 
CO 


^ 


CN 


00 
CM 


rH 


jq 


a 


CN 


00 


d 


00 


rH 
CM 


NO 


CO 


in 


o 


NO 


NO 


a;5 




-H 


1 ^ 


S83 


s 


g^ 


CO 


ON 


8 


s 


CO 


ON 


S 


s 


^ 


S 


8 


m 


8 


^ 


sa§8 


f 




3 


a 


J5; 


LTl 

rH 


^ 


;5^ 


S 


s 


r^ 


r~ 


ON 


CO 


<r 


d 


vO 


ON 


CO 


■^ 


r~ 


o 


;5 


m 

rH 


a?i 




?5 1— t 


:$ 


CO 
CO 


gi 


CM 
CM 


?8 


CO 
CO 


a 


CO 
CO 


-a- 

ON 


S8 


m 


d 


in 

CO 


CO 
CO 


8 


NO 

o 


CM 


CO 
CO 


!^ 


ON 
CO 


IC 


a8 


?1 


§ s 


00 


?R 


3 


?? 


^ 


CO 


;5 


ON 


rH 

eg 


^ 


S 


00 


3 


00 


o 


NO 


<f 


00 


^ 


r~ 


NO 


<f o 






Z rn 


§8 


s 


ON 


s 


K 


s 


S8 


a 


g 


^ 


ON 


s 


^ 


rH 


S 


00 


a 


ON 


g 


?i 


S 


as 






CO 


3 


aa 


^ 


J^ 


CM 

CM 


:l 


00 


'.O 


dP 


■<f 


a 


ON 


NO 


CO 


CO 


O 


NO 


iH 


a 


<y\ NO 




e 






W 



480 



1 § 

4-) 
(0 

a 
I 

i 

I 



CSlOOgg 



CM u~) r^ 
on 00 r^ 





(X) 
0^ 


^ 


rH 


(!3 




y 




1 




w 


S 


^ 


3 



00 



1^ CNJ 

.... .... 

OOOO OOOolO 



8 ^S^iSS ?iS88 SRf^fi8 
■ .... .... .... 

(— > --►«^i^n<^ ^oo~3- mr~.r^o 



-* CO m 00 



!nS^S ?J8^8 S?8!^g ai588 S :g ^ ?^ 

• ••• .... .... .... •••• 

<rcMCMCM cslOcMO '.DrO'-a--* tHOOO r^t^u~i>£) 

lOCMQro ooOCPiM t^ONf^o r^OOO OOr-^O 

oo>a-Oi-i cj\r^r~i-i oromcM coooO cnoomO 

Oi— loro cMrHrHm cMinoovo r-Iodd r-!vorod 

rorooorn ■:3"-3:C;<^ cJ^o^r^Q Lnv£lo^Q oor^oooo 

mmi^r-J cy,v3-mH L0<fcj\O a\vouio t'SrHcMc'S 

<ft-(rHcM rHOcMCM loromo OOOO ooa\oo^ 

S8S8 ^. SS8 f^. ^. S8 ^, f^. 3 8 S. ^. a 8 

OOcmO coOcmO ^cmcmO iHOCMLO r^oovoto 

rH-^iri-* ooQOQ Ovror^m r^ooocM rHOOcO-J 

uiOovO CTvOOO LnrOLO-<f ONinooo a^CMCMvo 

rocMoro phooo mvo<rm di-ndr-H oocjNinro 

a\aNQQ QSvor^ vO"^cMr^ sP^^-CQ OQcmq 

r^c^oo LnCT\\ou~i 'SvoHm O'SiTSo iHvSooo 



• • • • 

OOOO